Reintegrating young combatants in the Kivus What former child soldiers say

By Martin Sjøen Supervisor: Christian Webersik

May 2012

The master thesis is carried out as a part of the education at the University of Agder and is therefore approved as such. However, this does not imply that the University answers for the methods that are used or the conclusions that are drawn. Martin Sjøen Master Thesis University of Agder

Department of Development Studies Faculty of Economics and Social Sciences University of Agder

A conceptual study on the socioeconomic reintegration of former child soldiers in the Kivus, DRC

The above photo is of former child soldiers participating in an intervention program in the outskirts of Bukavu, South Kivu. It was taken by Ole Kåre Eide, and is reprinted here with kind permission.

i

Martin Sjøen Master Thesis University of Agder

Abstract The use of child soldiers to perform violent acts is one of the most condemned and emotionally disturbing practices globally. Child soldiers are with contextual variations believed to partake in most protracted social conflicts. International media, policy-makers and humanitarian actors have in recent years pushed the issue of child soldiering forward, making it one of the most recognized humanitarian issues. Yet the complexity of this issue is constrained by a lack of veritable knowledge. This thesis identifies common traits of former child soldiers who have reintegrated back to their families and local communities. Socioeconomic effects of formalized intervention efforts are also assessed. The empirical investigation is based on the testimonies of twenty-four former child soldiers in the Kivu provinces, DRC. Considerable challenges affect the reintegration process. Few former child soldiers are able to establish sufficient economic livelihoods. Social reintegration is a strenuous task with varying occurrences of prejudice, discrimination and stigmatization. Intervention programmes seem incapable of addressing these challenges in a sufficient way. Nonetheless, child soldiers express considerable satisfaction with intervention efforts. The overall process of reintegrating former child soldiers is surrounded by ambiguity. Conventional approaches are preoccupied with assessing established socioeconomic variables. Contextual factors like family and social roles are important aspects related to a successful reintegration.

Key words: Democratic Republic of Congo, Kivus, child soldiers, intervention efforts, socioeconomic reintegration.

ii

Martin Sjøen Master Thesis University of Agder

Acknowledgements

Many people have contributed towards this work in a great number of ways. I would like to start by thanking Aase and Thor Magne Kommedal for their invaluable assistance and hospitality. Their compassion and dedication serve as an inspiration to me and many alike. Secondly, I owe my deepest gratitude to Christian Webersik who supervised and supported me throughout this work. I am forever grateful for your insight and guidance. Further, I am indebted for the encouragement and endorsement provided by Professor Arne Olav Øyhus. I would also like to show my gratefulness to the entire staff at the Centre for Development Studies, my fellow students, and previous cohorts at the University of Agder for their inspiration, friendship and support. The fieldwork could not have been completed without a number of helpful organizations and institutes. It is an honor for me to thank the Pentecostal Foreign Mission of Norway (PYM), Christian Relief Network (CRN), the RAFTO Foundation, Centre for Intercultural Communication (SIK), and the German Development Agency (GIZ). Lastly, I would like to thank all the respondents in DRC. Your testimonies are the foundation of this work, and I thank you all for your participation. This thesis is written in a sincere effort to understand and contribute towards the peacebuilding DRC.

iii

Martin Sjøen Master Thesis University of Agder

Table of content Abstract ...... ii Acknowledgements ...... iii List of figures ...... vii List of tables ...... vii List of appendices ...... vii Abbreviations and acronyms ...... viii

Chapter 1: Introduction ...... 1 1.1. Background ...... 1 1.2. Study area ...... 2 1.3. Research objectives ...... 3 1.4. Research questions ...... 3 1.5. Methodology in brief ...... 4 1.6. Motivation for choice of topic ...... 4 1.7. Thesis outline ...... 4

Chapter 2: Theoretical background ...... 5 2.1. Child soldiers ...... 5 2.1.1. Defining child soldiers ...... 5 2.1.2. Child soldiering, a contemporary phenomenon? ...... 7 2.1.3. The global use of child soldiers ...... 8 2.1.4. Increased international attention ...... 10 2.2. Causes of child soldiering ...... 11 2.2.1. Supply and demand for child soldiers ...... 11 2.2.2. Contemporary social protracted conflict ...... 13 2.2.3. Active agents or victims? ...... 14 2.3. The impacts of child soldiering ...... 16 2.4. Child soldier intervention ...... 18 2.4.1. Defining DDR-(RR) ...... 18 2.4.2. Assessing child soldier intervention ...... 19 2.4.3. The concept of reintegration ...... 21 2.4.4. Reintegration framework ...... 22 2.5. A cursory overview of trends in literature ...... 24 2.5.1. Notable research gaps ...... 25 2.6. Contextual background ...... 26 2.6.1. Brief political history ...... 26 2.6.2. Historicity of the Congolese Wars ...... 28 2.6.3. First Congolese War (1996-1997) ...... 28 2.6.4. Second Congolese War (1998-2003) ...... 29

iv

Martin Sjøen Master Thesis University of Agder

2.6.5. Sub-conflicts of the Congolese Wars (2003-Present) ...... 30 2.6.6. Child soldiers in DRC ...... 31 2.6.7. Choice of sites ...... 33 2.7. Theoretical framework ...... 36 2.7.1. Applying the theoretical framework ...... 37

Chapter 3: Methodology ...... 39 3.1. Research strategy ...... 39 3.1.1. Research design ...... 39 3.1.2. Participatory research ...... 40 3.2. Sampling ...... 41 3.3. Data collection ...... 42 3.3.1. Interviews ...... 42 3.3.2. Focus groups ...... 43 3.3.3. Participant observation ...... 44 3.4. Interpretation of data ...... 45 3.5. Limitations...... 45 3.6. Ethical considerations ...... 46

Chapter 4: Findings and analysis ...... 48 4.1. Disposition...... 48

Part 1 ...... 49 4.2. Background information ...... 49 4.2.1. Age dispersal ...... 49 4.2.2. The circumstances of recruitment ...... 50 4.2.3. Drivers of becoming a child soldier ...... 52 4.2.4. Child soldiers' roles ...... 53 4.2.5. Departing the armed groups ...... 55 4.2.6. Re-recruitment ...... 56 4.2.7. Summary of main findings part one ...... 57

Part 2 ...... 58 4.3. Socioeconomic reintegration ...... 58 4.3.1. Cultural barriers ...... 58 4.3.2. Family reunification ...... 58 4.3.3. Family acceptance ...... 59 4.3.4. Active agency vs. victimhood – family reintegration ...... 62 4.3.5. Community reintegration - child soldiers’ perspective ...... 63 4.3.6. Community acceptance ...... 64

v

Martin Sjøen Master Thesis University of Agder

4.3.7. Addressing communal inequity ...... 65 4.3.8. Community perspective on child soldier reintegration ...... 66 4.4. Conceptualizing hostility ...... 68 4.4.1. Hostile behaviour among former child soldiers ...... 69 4.4.2. How local communities perceive former child soldiers ...... 70 4.4.3. Hostility towards former child soldiers ...... 71 4.5. Economic livelihood ...... 72 4.5.1. Economic impact of armed conflict on child soldiers ...... 74 4.6. Psychological well-being...... 74 4.6.1. Exposure to distress factors ...... 75 4.6.2. Symptoms of psychological distress...... 76 4.6.3. Received psycho-social support ...... 77 4.7. Summary of main findings part two ...... 79

Part 3 ...... 81 4.8. Assessing intervention efforts in the Kivus ...... 81 4.8.1. Intervention efforts towards social reintegration ...... 81 4.8.2. Intervention efforts towards education and economic restoration ...... 84 4.8.3. Contextual dimensions towards intervention work ...... 87 4.8.4. Key patterns of intervention efforts ...... 89 4.9. Peacebuilding through conflict transformation ...... 91 4.9.1. Destructive conflict ...... 92 4.9.2. Consequence of destructive conflict ...... 94 4.9.3. Constructive conflict transformation ...... 94 4.9.4. Consequence of constructive conflict transformation ...... 97 4.9.5. Summary of main findings part three ...... 98

Chapter 5: Conclusion ...... 100 5.1. Contextual child soldier recruitment, usage and departure ...... 100 5.2. Conjecturing common traits of former child soldiers ...... 102 5.3. Reasoning the effects of intervention efforts ...... 103

References ...... 104 Appendices ...... 119

vi

Martin Sjøen Master Thesis University of Agder

List of figures Figure 1: Map of eastern DRC 2 Figure 2: Global distribution of child soldiers 9 Figure 3: Vulnerability factors framework explaining child soldier recruitment 12 Figure 4: Defining the DDR-(RR) process 18 Figure 5: Map of DRC 26 Figure 6: Map of study sites 34 Figure 7: Theoretical frameworks model 38 Figure 8: Diagram on the mode of recruitment 51 Figure 9: Dispersal of armed groups 52 Figure 10: Diagram on the roles of child soldiers 54 Figure 11: Diagram on the modes of departing armed groups 56 Figure 12: Diagram on the dispersal of occupations 73 Figure 13: Map of northeastern DRC 75 Figure 14: Diagram on psycho-social support 77 Figure 15: Diagram on perception of intervention efforts 88 Figure 16: Course of conflict 91 Figure 17: Diagram on social acceptance 95

List of tables Table 1: Age dispersal among the respondents 49 Table 2: Table on social relationships with family and relatives 61 Table 3: Table on community reintegration 66 Table 4: Table on traumatic exposure 78

List of appendices Appendix 1: Interview guide: In-depth interviews 120 Appendix 2: Overview of respondents: Background information 124 Appendix 3: Overview of collaborating organizations and institutes 125 Appendix 4: Ethical guidelines 126

vii

Martin Sjøen Master Thesis University of Agder

Abbreviations and acronyms

ADF Allied Democratic Forces (Uganda rebel group) ADFL Alliance des Forces Démocratiques pour la Libèration du Congo-Zaire (Alliance of Democratic Forces for the Liberation of Congo-Zaire) AFG Armed Forces and Groups CAFF Children associated with fighting forces CBO Community based organization CNDP Congres National pour la defense du Peuple (Congolese Tutsi rebel group) CRC Convention on the Rights of the Child CRN Christian Relief Network DAI Development Alternatives Incorporated DDR-(RR) Disarmament Demobilisation and Reintegration (Rehabilitation and Repatriation/Resettlement) DRC The Democratic Republic of Congo FAC Forces Armées Congolaises (Congolese Regular Armed Forces between 1997- 2002) FAR Forces Armées Rwandaises (Rwandan Regular Armed Forces) FARDC Forces Armées de la Republique du Congo (Congolese Regular Armed Forces) FAZ Forces Armées Zairoises (Zairian Regular Armed Forces up to 1997) FDLR Forces Démocratiques de Libération de Rwanda and Rasta (ex-Far and Interahamwe soldiers) FTS Free the Slaves GIZ Deutsche Gesellschaft für Internationale Zusammenarbeit (German Development Agency) IASC Inter-Agency Standing Committee ILO International Labour Office IRC International Rescue Committee LRA Lord's Resistance Army LTTE Liberation Tigers of Tamil Eelam MDRP Multi-country Demobilization and Reintegration Program MLC Mouvement de Liberation Congolais (Movement for the Liberation of Congo) MONUC Mission d'Organisation Nations Unis au Congo (United Nations Mission to the Congo 2000-2010)

viii

Martin Sjøen Master Thesis University of Agder

MONUSCO Mission de l'Organisation des Nations Unies en République démocratique du Congo (United Nations Mission to the Congo from 2010) NGO Non-governmental organization NUPI Norwegian Institute of International Affairs PSC Protracted social conflicts PTSD Post-traumatic stress disorder PYM Pentecostal Foreign Mission of Norway RCD Rassemblement Congolais pour la Démocratie (Congolese Rally for Democracy) RENAMO Resistência Nacional Moçambicana (The Mozambican National Resistance) RPF Front Patriotique Rwandais (Rwandan Patriotic Front) SADC Southern Africa Development Community SGBV Sexual and gender based violence SIK Centre for Intercultural Communication SWAY The survey of war affected youth UN United Nations UNICEF United Nations Children's Fund UNITA União Nacional para a Independência Total de Angola (National Union for the Total Independence of Angola) UNDP United Nations Development Programme

ix

Martin Sjøen Master Thesis University of Agder

Chapter 1: Introduction 1.1. Background The forcible use of child soldiers to perform organized killings is perhaps one of the most heinous known human atrocities occurring today. Yet it is believed that child soldiers are actively involved in most protracted social conflicts (PSC) globally (Brett and Specht 2004: 1). Increased attention has in recent years pushed the agenda of child welfare forward, often placing it at the forefront of much humanitarian and policy-work (Andvig 2006: 7, Beber and Blattman 2011: 1). Still, many leading experts are requesting more scholarly research on child soldiering, arguing that the field is in great need for further veritable knowledge (Andvig 2006, Gates and Reich 2010, Hart and Tyrer 2006, Singer 2006). Most of what is written on the phenomenon of child soldiering deals usually with its coercive use and victimization of children in conflict (Brett and Specht 2004: 1). Popular images of youth deprived of a peaceful existence, is often reinforced by advocacy groups “to attract attention and gain support” (Hart and Tyrer 2006: 9). Long-term outcomes of child soldiering, both for young soldiers and their surrounding environments, have largely been overlooked until recently. The transition of reintegrating child soldiers back to a civilian life, suffers especially from this lack of scholarly interest. Some authors argue that contextual appropriate intervention programmes, can be sustainable tools for aiding former child soldiers back to a civilian life (Singer 2006: 211, Wessells 2006: 155). Yet it is also argued that scarcity of scholarly knowledge may nonetheless implicate the construction of such appropriate programmes (Wessells 2006: Preface). As a result, many child soldier intervention programmes have been trial and error based, developed out of “myths and assumptions” with immensely varied success (Blattman and Annan 2008, Singer 2006). The Democratic Republic of Congo (DRC) provides us with a current example of a country that has experienced a great number of children affected both as perpetrators and victims of violent conflict (Agborsangaya 2000: 1). The devastating effect of violent conflict in DRC is perhaps unprecedented in recent years, leaving millions of injuries, casualties and displaced (IRC 2007). This work will address some of the research gaps on the process of reintegrating child soldiers back to a civilian life in DRC. There is a great need to understand the phenomenon of child soldiering including the aspect of formalized intervention support. This is especially important within the contextual circumstances of a war-torn Congo.

1

Martin Sjøen Master Thesis University of Agder

1.2. Study area Being among the largest geographical countries in the world, DRC lies centrally located in “tropical” Africa. This research has been conducted in collaboration with different organizations1 in the Kivus (North and South Kivu provinces). Located in northeastern DRC, the Kivus has been the epicentre of the armed conflict since 1998 (IRC 2007: Preface). Figure 1: Map of eastern DRC. Child soldiering is a long standing Source: BBC (2010). phenomenon within DRC, but its comprehensive use emerged in the mid-1990s, when armed groups2 initiated widespread child soldier recruitment. As the aftermaths of the Rwandan genocide moved across the Congolese border in 1994, increased ethnic and communal violence spread in the northeastern region, eventually culminating in a regional war (Dunn 2002: 55, Murison 2002: 226). Often described as the “World War of Africa”, the First Congolese War (1996-1997) was followed by the subsequent Second Congolese War (1998- 2003). At least eleven African countries and numerous armed groups have been involved in the fighting’s since 1996 (Trefon 2011: xv). Despite many formalized peace agreements, revival of violence (sub-conflicts of the second Congo war), and widespread human-rights abuse is an on-going problem (Security Council 2010). Since the mid-1990s, millions of Congolese have periodically been internally and externally displaced (Murison 2002: 228, Trefon 2011: 19). According to United Nations (UN), child soldier recruitment is still pervasive in DRC (Tonheim 2011: 4). In addition, malnutrition, diseases and inadequate healthcare has led to a regressive human development (Britannica 2011, UNDP 2011). The International Rescue Committee estimates that the contemporary armed conflict in DRC have caused some 5.4 million “excess deaths” (IRC 2007: ii). While this mortality rate is commonly cited by many scholars, it is also surrounded by uncertainty. Demographers Lambert and Lohlé-Tart for example, question these figures arguing that the overall number is in actuality considerably lower (2008). Yet there remains little controversy in the statement that DRC has been the centre of much human suffering in recent years.

1See Appendix 4 for an overview of different collaborating organizations and institutes. 2While the author acknowledges the difference between phrases like “armed groups”, “rebel groups”, “armed forces”, “warring groups” and “militia groups”, “armed groups” will be consistently used as a collective term for the mentioned throughout this paper. 2

Martin Sjøen Master Thesis University of Agder

1.3. Research objectives The objective of this research is to examine the child soldier reintegration process in the Kivus, DRC. A special emphasis will be to investigate the effects of child soldier intervention programmes. Basic social3 and economic4 variables are commonly identified as being among the most important factors in child soldier reintegration (Verhey 2001: 22), and the examination of these will be at the heart of the thesis.

Sub-objectives:

1. Conducting participatory research to contribute towards the research gaps on child soldiering in northeastern DRC.

2. Identifying challenges faced by former child soldiers, their families and local/host communities when the former are being reintegrated into the latter.

3. Assessing the effects of formalized intervention programmes.

1.4. Research questions In order to examine important aspects of the child soldier reintegration process in the Kivus, DRC, these specific research questions have been formulated:

1. What are the common socioeconomic traits of former child soldiers when being reintegrated back to their families and local or host communities?

2. What are the key patterns of formalized child soldier intervention support, and how are these efforts impacting the socioeconomic recovery process for former child soldiers?

3. How are formalized interventions programmes affecting conflict transformation efforts in the Kivus?

3Family reunification, alternative placement and community reintegration. 4Basic and higher education, vocational training and income generating work. 3

Martin Sjøen Master Thesis University of Agder

1.5. Methodology in brief This research relied on a qualitative approach for investigating its objectives. Qualitative strategy fits well in this kind of work according to Bryman (2008: 22), as the main purpose is to emphasize the conveyed words of participants. Primary data was collected in northeastern DRC from mid-January to mid-February 2012. Research design is case study and multiple methods approach was applied. These methods were in depth interviews, focus groups and observations.

1.6. Motivation for choice of topic The desire to conduct such sensitive and challenging research lies in a great interest in peace and conflict studies. Another decisive factor has been a strong belief in the concept of conflict transformation. This concept which will be further elaborated in the theoretical framework, yet it emphasizes the conflict-in-context, meaning that social and regional considerations must be made in facilitating and managing sustainable peace (Miall 2004: 2). Several years of working experience with vulnerable children in humanitarian organizations and psychiatry, has further fueled the commitment for conflict transformational work. Still, the strongest rationale for undertaking such a project lies in a firm belief that people affected by adversity, can and should play a decisive role as active agents when addressing issues concerning their own well-being.

1.7. Thesis outline The thesis is organized as following: Chapter 1 provides a brief introduction, background on the study area, methodology, research objectives and the justification for the choice of topic. Chapter 2 clarifies concepts, revisits academic literature, gives a historic overview of the study area, depicts the sites of investigation and presents the theoretical framework. Chapter 3 portrays the methodology used and describes the research process. Chapter 4 presents the empirical findings in three parts, subsequently discussing them in the light of the conceptual and theoretical framework. Chapter 5 provides the concluding remarks.

4

Martin Sjøen Master Thesis University of Agder

Chapter 2: Theoretical background 2.1. Child soldiers

2.1.1. Defining child soldiers A child soldier is any person under 18 years of age who is part of any kind of regular or irregular armed force or armed group in any capacity, including but not limited to cooks, porters, messengers and anyone accompanying such groups, other than family members. The definition includes girls recruited for sexual purposes and for forced marriage. It does not, therefore, only refer to a child who is carrying or has carried arms. (Cape Town Principles 1997: 12).

There is a vast amount of definitions on the term “child soldiers”, each aiming to capture different nuances and roles served by young humans affiliated with armed groups5. Whilst most derive from an age-based understanding, normative aspects are also included in some definitions. International Labour Office (ILO) for example, describes the forced recruitment of child soldiers as one of the “worst forms” of child labour in the world (2003: i). Kevin Bales on his part categorizes forced recruitment of child soldiers as the “worst kind” of modern slavery6. The introductory quote from the Cape Town Principles is perhaps the most recognizable to date. According to Andvig (2006: 12), it is also the basis for most available statistics in the field. A certified strength with this age-based definition is surely the theoretical ease of which it can be applied around the world. It does however offer some practical challenges in return. Perhaps most prominent is the contested nature of categorizing any person regardless their individual background, merely on their age. Some authors have questioned the credibility of such a definition, arguing that asserting “childhood” to a specific age is both vague and artificial (Andvig and Gates 2010: 2, Boyden and de Berry 2007: xxi, Shepler 2004: 2). Constructionism as a post-modern theory views social phenomena like childhood as artefacts constructed within a specific context. What constitute a child varies between cultures, thus childhood can be considered a social construct (Hart and Tyrer 2006: 6). Brett and Specht further points out the challenge of direct comparison between Western and non-Western settings, as the expectations towards children and the roles they serve are often radically different between the two settings (2004: 1). This point finds relevance among other authors who claim that children in non-Western settings, are expected to make “mature decisions” and partake in “adult activities” at an earlier age than their Western peers (Kostelny

5See for example African Charter 1999, ILO Convention No.182 1999, UN Convention Optional Protocol 2000 and Paris Principles 2007. 6Bales, K. President, Free the Slaves (FTS). Conversation: 04.11.2011. 5

Martin Sjøen Master Thesis University of Agder

2006:20, Shepler 2004: 12, Singer 2006: 63). Leading discourses have until recently been predominantly liberal and Western based, often at the expense of overlooking contextual factors from other non-Western settings (Schmidt 2007: 57, Wessells 2006: Preface). Another implication stems from the inaccuracy of the two words that make up the term “child soldier”. Most underage soldiers are likely not children per se; rather it is the adolescents who account for the vast majority of child soldiers globally. Yet according to a UN study, adolescents receive the least amount of attention in armed conflict, thus becoming one of the most neglected groups in war (Brett and Specht 2004: 3). Adolescence is equivalent to childhood a rather vague term, and the boundaries between these phases are unclear. Traditionally it is understood as the transition between childhood and adulthood, sometimes referred to as the “teenage” phase with regional and cultural variations (Tefferi 2010: 24). The superiority of the child soldiers is namely recruited in their teens, due largely to their relative physical and cognitive development (Hart 2008: 280, Psychology Beyond Borders 2008: 10). Few people would arguably oppose the distinct difference between a six year old and a seventeen year old, both in terms of maturity and capability (Andvig and Gates 2010: 2). Child soldiering does not become any less illegal or even morally acceptable with the vast majority of underage soldiers being teenagers and not children (Singer 2004: 574). But in the words of the Norwegian researcher Jens Chr. Andvig, perhaps it would be more suitable to discuss child soldiering as an “issue related to early adolescence than a child issue proper” (2006: 13). It has also been postulated that categorization “soldier” is not entirely representative, chiefly since it emphasizes those who actually served the specific role as active soldier (Psychology Beyond Borders 2008: 8). There is great uncertainty regarding the tasks and roles attributed to child soldiers, which may suggest that it is case and context dependent (Wessells 2006: 8). Some authors propose that many if not most child soldiers are perhaps not “active” soldiers, rather they perform other non-combatant duties within the different armed groups (Andvig 2006, Andvig and Gates 2010, Gates and Reich 2009). A statement which should be noted is the scene of much debate (Singer 2006: 76-77). There are studies which suggest that a high number of child soldiers have actively participated in combat in Uganda and DRC (Bayer et al. 2007, ILO 2003), Sri Lanka (Kanagaratnam 2005) and Liberia (Sendabo 2004). Some studies however, do not specify whether active fighting was events of isolation or the main occupation for the child soldiers. Contrasting views are put forward by other authors who imply that the main roles for most child soldiers might be related to doing “household” tasks (Andvig and Gates 2010: 2). Regardless, by defining them as soldiers, the

6

Martin Sjøen Master Thesis University of Agder risk of overlooking children who have contributed in other ways and girls in particular is great. Until very recently, girls have only occasionally been recognized as child soldiers both in academic work and post-conflict efforts (Tonheim 2010: 18, Wessells 2010: 194). These imputations have prompted the creation of alternative terminologies, perhaps most noticeable, the “children formerly associated with armed forces and groups” (AFG), and “children associated with fighting forces” (CAFF), (Paris Principles 2007: 7, Psychology Beyond Borders 2008: 8). Nevertheless, most amended definitions are in the same manner criticized, as the potential for neglecting or accentuating certain groups of children and adolescents is ever so present (Wessells 2006: 6). It is a paradox then that most of the today’s child soldiers are likely not children per definition, or perhaps not even “soldiers” in a traditional sense. A rigorous discussion on the appropriateness of defining child soldiers based on their served role or age is certainly justified, yet it does not comply within the objectives of this thesis. Tonheim rightly argues that “the need to draw an age limit stems from inter- national and national legal considerations, and a focus on age is therefore both inevitable and necessary” (2010: 16). Moreover, Andvig states that neither a cultural, psychological, economical or any other contextual appropriate definition would be as operational as the “age” definition (2006: 12- 13). While this author acknowledges the many limitations and possible implications of using the term “child soldier” as initially defined in this section (Cape Town Principles), for readability, it will nonetheless be consistently used throughout this paper.

2.1.2. Child soldiering, a contemporary phenomenon? There is an inherent debate whether child soldiering is a historic or modern phenomenon. This section will surely not contribute towards a closure to that debate; instead it will aim to highlight some of the prevailing arguments in it. In a short but thought provocative article featured in Foreign Policy, Gates and Reich (2009) encapsulates important aspects as they dismantle scores of likely misconceptions on child soldiering. Among the most common misconstructions is possibly the idea that child soldiering is a recent post-Cold War phenomenon involving a “pre-adolescent African boy, perhaps doped, wielding an AK-47 with anger burning in his eyes” (Gates and Reich 2009). Historians of warfare would in all probability argue that children and adolescents have been involved in armed conflict for most of our conceivable past (Lee 2009: 4, Schmidt 2007: 55). Yet, the scale of child soldier usage historically is a contested subject. Case in point; Singer states that in European history at least, children as a “general rule” were excluded from direct military service and roles related to

7

Martin Sjøen Master Thesis University of Agder active fighting (2006: 11). They could however occasionally be found serving other auxiliary roles. What constitutes the “general rule” is a bit unclear, though it is believably tied with the role as a soldier in a conscripted or regular army, thus also a valid point. However, in present context, the largest child soldier recruiters are perhaps also among irregular and non-state actors (Gates and Reich 2010: 4), though there are believably large numbers of children in conventional armies around the world (Schmidt 2007: 49). Firmly grounded in the controversial belief that many child soldiers rarely or never experience active fighting, contemporary and historic child soldiering may in fact not differ that radically from one another. Furthermore, last century did in actuality witness a considerable usage of conscripted child soldiers on European soil. Thousands of underage British soldiers fought during World War I (Lee 2009: 3). Their German counterpart would a few decades later be put to use in plentitude, as the deteriorating Third Reich deployed the Hitler Youth as a last act of strategic desperation (Singer 2006: 14). These are just a few of many examples of child soldiering in a Western context. A broad but noteworthy difference between the young men participating in the Great Wars and today’s young soldiers, is that the former are sometimes portrayed as “heroes” while the latter are “victims” (Lee 2009: 3). Child soldiering may rightly have been “unevenly” distributed throughout the history of warfare, but evidence seems also to suggest that children have been utilized in armed conflict when the circumstances, such as need and availability have been “right”. Regardless, Singer does make a compelling argument on how children's affiliation with armed groups has changed in accordance with contemporary protracted social conflicts (2006: 15). Especially when considering how warfare has evolved from something that in historic perspective mainly involved regular armies, to present day where it embraces every aspect of society, and civilians suffers the greatest casualties (Barash and Webel 2009: 18, Singer 2006: 4-5). While reliable statistics are difficult to obtain, children may constitute 10% of the world’s combatants, involved in 75% of the conflicts (Singer 2006: 29-30). In some conflicts, children are in large numbers immersed in direct fighting’s, often committing harrowing acts (Wessells 2006: 3). It is therefore difficult to blatantly refuse that child soldiering has not altered over time, though surely not in a positive manner.

2.1.3. The global use of child soldiers The phenomenon of child soldiering is a global issue, with children participating in a great number of armed conflicts on all continents. A figure of 300,000 is commonly stated to

8

Martin Sjøen Master Thesis University of Agder be the total number of child soldiers globally, yet this is a disputed subject. Gates and Reich (2009) claim that this figure was an educated guess put forward by advocacy groups in the 1990s, and even if it was ever true, it would not be now. Several armed conflicts have ended since then, many of which used child soldiers to a great extent as in Angola, Liberia, Nepal and Sri Lanka. Some authors have scaled down the global estimate of child soldiers from approximately 100,000 (Andvig 2006: 8), to 200-250,000 (Gates and Reich 2010: 10,251). Michael Wessells agrees that these estimates should be taken with a “grain of salt”, yet he insists that the 300,000 is relatively accurate, as it is constructed out of many careful country studies (2006: 9). In addition, there may be approximately 500,000 children among armed groups currently not involved in violent conflict (Singer 2010: 94). The picture below illustrates the dispersal of child soldiering globally. Child soldier usage is more widespread in countries marked with red.

Figure 2: Global distribution of child soldiers. Source: United Nations (2011). Andvig boldly states that even if the total number of child soldiers is 300,000, this figure is nonetheless modest compared to children participating in other “adult” economic activities (2006: 9). Some estimates suggests that there are approximately 250 million children involved in daily labour, whereas 50 million of them in the “worst forms”, which essentially child soldiering is (Andvig 2006: 9). The overall figure of child soldiers is also nominal compared to the number of non-combatant children severely affected by armed conflict (Gates and Reich 2010: 5). More than two million children have been killed in war during the last decade (Heart and Tyrer 2006: 4). Non-combatant children are believed to suffer the most casualties in armed conflict. According to Singer, one out of every two-

9

Martin Sjøen Master Thesis University of Agder hundred child globally is impaired with “war-related psychological malady” (2010: 98). The overall number of child soldiers has by some authors therefore been argued to be neither as interesting nor as important, as understanding the variations between armed groups, regions and countries in terms of child soldier recruitment (Gates and Reich 2009, Wessells 2006: 12). With the end of several conflicts in Asia, Africa is generally regarded as the epicenter of child soldier usage (Achvarina and Reich 2010: 56, Singer 2006: 19). As illustrated in the introductory quote, child soldiering involves a magnitude of roles for the children to partake in. Some armed groups deploy children directly into active combat, while others make use of the underage soldiers in more complementary roles (Gates and Reich 2010: 3). Previous assertions often dismissed girls as being active child soldiers, yet more up to date research have since renounced this. Many girls partake in active fighting comparable to their male counterparts, and sometimes in proportionate numbers. Nevertheless, female child soldiers are often associated with more sexual exploitation than males (Wessells 2006: 8). Girls may comprise up to 40% of the world's total number child soldiers (Psychology Beyond Borders 2008: 10).

2.1.4. Increased international attention So what has made the subject of child soldiering a global phenomenon? Some researchers claim that the principle change is our increased awareness on the issue, alongside the contemporary Western perceptions on childhood being a “sacred time reserved for innocence, learning, and play” (Gates and Reich 2009, Hart and Tyrer 2006). Andvig concurs as he argues that child soldiering has become a “global issue because of the feelings it awakes, not its quantitative significance” (2006: 7). The Iran-Iraq War that transpired in the 1980s, has been described as an international turning point, when it attracted significant attention to the immense use of child soldiers by both countries (Heckel 2005: 12, ILO 2003: 2). The first global study on the impacts of armed conflict on children was then put forward by the United Nations in 1996 (Wessells 2006: 9). Published by Graça Machel (1996), it is considered to have established the foundation of much international child protection agenda ever since (Gates and Reich 2010: 3). Shortly after, the United Nations General Assembly adopted the Convention on the Rights of the Child (CRC), which came into force in 1990. In the following years, children in armed conflict has been one of the most recognized and emotional issues among humanitarian organizations, media, international institutions and policymakers around the world (Gates and Reich 2010: 3). In that space of time, a vast number of legal provisions and protocols has been created, while substantial practical efforts

10

Martin Sjøen Master Thesis University of Agder has been made to try and address the issue of child soldiering (Druba 2002: 272, Singer 2004: 568). Corresponding with the fundamental disagreement on whether child soldiering is a “new” or an “age old” phenomenon. Leading authors also differ on whether the international normative and legislative frameworks on child soldiering have been “remarkable” or “limited” successful. Beber and Blattman argue that the prosecution of high-profiled child soldier recruiters, a global reduction of state recruitment and negotiation with armed groups are noticeable signs of success (2011: 1). Gates and Reich on the other hand, claim there is little evidence that the strategy of “naming and shaming” has been effective as child soldier recruitment is a persistent problem (2010: 4).

2.2. Causes of child soldiering Turning away from this brief conceptual discussion on the historic development of child soldiering, the attention will now be directed towards more comprehensive perspectives on why some children become involved in armed conflict. Much has been written on what causes child soldier recruitment. Extensive work has focused on broad push/pull factors when describing how some children become part of armed groups (Wessells 2006: 46). The changing nature of warfare, the breakdown of ancient social and familial structures, or the notion of “failed states” is also commonly cited as leading theories (Singer 2006: 96). Other perspectives include the examination of forced vs. voluntary recruitment. It is now commonly accepted that child soldiering is not grounded in specific social contexts or “barbaric cultures”, illustrated in the very uneven distribution of child soldiers globally (Andvig and Gates 2010: 88, Gates and Reich 2010: 9). Brett and Specht assert that few things in life are so “clear-cut that there is one single explanation for them” (2004: 9).

2.2.1. Supply and demand for child soldiers Secure in the knowledge that the nature of recruitment is highly volatile, what are the factors that determine the need and availability for child soldiers? Andvig and Gates argue that most research on child soldier recruitment has been focusing on the supply side, “i.e. the number of children available for recruitment” (2010: 16). Nonetheless, more recent work has postulated that the scope of child soldier recruitment and also the welfare of the child soldiers, is dependent more on the armed group (demand), than contextual factors such as availability (supply) (Andvig 2006: 7). Hence, a key to develop a stronger understanding and even prevent child soldiering, is perhaps tied more to understanding first and foremost the characteristics of the fighting groups, than the characteristics of the area plagued by violent

11

Martin Sjøen Master Thesis University of Agder conflict (Andvig 2006: 17). This assertion correlates well with an argument from Singer who claims that “Child soldiering derives from a set of deliberate choices from military leaders” (2006: 135). These choices may stem from a belief that children are more cost effective and easier to manipulate than adults, the almost abundant access to children in poor countries, their “shock” value, local practice, or the share dedication and effort put forward by child soldiers in service (Andvig and Gates 2010, Singer 2006, Somasundaram 2002, Wessells 2006).

Figure 3: Vulnerability factors framework explaining child soldier recruitment.

Education and work Culture and Family and tradition friends

Politics and Features of ideology adolescence

Push/pull factors for child Poverty Armed conflict soldier recruitment

In a more traditional approach, Brett and Specht designed an environmental factors framework to explain why some children and adolescents join armed groups (2004: 9). There are as they claim seven broad and interlinked factors that are “necessary but not sufficient”, to elucidate child soldier recruitment.

War is the most “crucial” environmental factor in explaining child soldiering, but in isolation, it does very little to elucidate why some children become soldiers and other do not (Brett and Specht 2004: 36). It has therefore been argued that these environmental factors are cumulative, meaning that the combination of certain socioeconomic factors will increase the chance of a child becoming a soldier. Poverty is often referred to as the one thing most child soldiers have in common (Singer 2006: 61-62). This factor is also attributed to explain why 12

Martin Sjøen Master Thesis University of Agder some violent conflict arise in the first place (Andvig 2006: 40, Collier 2008: 19). Yet much like the factor of war, poverty does very little to explain why some poor uneducated children become soldiers, when most do not. At the same time, it can hardly be discarded as an important reason for child soldier recruitment seeing that some many child soldiers have poverty in “common” (Brett and Specht 2004: 36). Evidently there is little consensus regarding the prevailing causal factors in turning children to soldiers, if one overlooks the existence of violent conflict. Andvig's claim regarding the characteristics of the armed groups being more important than the conflict-in-context is certainly novel. However, there is also reason to believe that the contextual circumstances of violent conflict will affect the armed groups, and in turn also impact their approach towards child soldier recruitment. In their regional studies of African civil wars, Achvarina and Reich (2004 and 2010) found strong correlation between poorly protected refugee-camps, and forced child soldier recruitment. What is more, these disaggregated studies found only small correlation between poverty, education, household assets and child soldier recruitment (Achvarina et al. 2008: 21). Surely this would imply that there is a great potential to further explore the relationship between child soldiering and different independent variables, in order to gain more verifiable knowledge.

2.2.2. Contemporary social protracted conflict The contemporary use of child soldiers has been argued to be a by-product of recent changes in armed conflict. Singer claims that 1) social disruption and failure of development, 2) a more brutal kind of conflict, and 3) technological improvements in weaponry, are the main factors for explaining this development (2006: 38). Social disruption can be understood as erasing the “ancient distinction” between combatant and civilian (Singer 2006: 4). In the course of last century or so, the percentage of civilian casualties in armed conflict increased from approximately 5% to 90% (Hart and Tyrer 2006: 4). As mentioned, violent conflict has transpired from taking place in relatively uninhabited areas mainly involving regular armies, to be at the very heart of civilian societies (Barash and Webel 2009: 18). More so, this is interlinked with the apparent brutality that has established itself as an almost norm in many contemporary social protracted conflicts (Gates and Reich 2010: 8). Wessells calls this the “new face of war” (2006: 18). On top of the breakdown of the “ancient” moral codes in warfare, more sinister means are used in armed conflict. Perhaps most noticeable, the widespread existence of sexual and gender based violence (SGBV) as a “weapon” (Singer 2006: 104).

13

Martin Sjøen Master Thesis University of Agder

Furthermore, the proliferation of light weapons since the end of the Cold War provides the possibility to arm children according to their physical stature, and it is a reasonable explanation for why some children are able to actively partake in violent fighting (Dallaire 2011: 12, Singer 2006: 38). There are an estimated 500 million handheld assault weapons globally, many available for very little money (Renner 1999 in Wessells 2006: 18). The profusion of such weapons has enabled armed groups to expand their military strength at a very rapid speed. Some authors question the legitimacy of this theory though, as it would indicate that children and adults are substitutes for one another within the structures of armed groups (Andvig and Gates 2010: 2-3). Child soldiering at least within the social sciences has predominantly been explored through theories of child labour. A dominant view here is that children and adults are employed to perform work that is complementary to each other, rather than substitutive (Andvig 2006: 7). This disagreement is presumably linked with the conflicting perceptions on the roles that child soldiers perform within their groups.

2.2.3. Active agents or victims? Another disputed topic related to child soldier recruitment, is whether children are mainly the victims of coercive enlistment, or if they can have any form of initiative in this involvement. This debate challenges some fundamental liberal perspectives, as the dominant perception of childhood is often associated with a time of vulnerability and innocence (Schmidt 2007: 57). Recent years has seen a growing recognition for the possibility of children playing a decisive role in their own development through participation or “active agency”. Child participation is also closely linked with child rights work (Hart and Tyrer 2006: 13, Wessells 2006: 228). However, this perspective has largely been absent in the context of children in armed conflict. Hart and Tyrer relates this shortcoming to three distinct factors, 1) a child's need for protection from violent conflict, 2) the socially constructed Western notion of childhood, and 3) liberal humanitarian images of children in conflict (2006: 9). These predetermined ideas of childhood may ultimately implicate the complex process of child soldiering. Hart further argues that child soldier recruitment should be viewed less through a humanitarian scope and “more in relation to political-economic processes” (2008: 277). It might be difficult to imagine anyone, let alone children, volunteering to be a part of what are essentially groups committing violent acts. Yet when keeping in mind two previous discussions in this paper; 1) Childhood as a social construct and, 2) the pervasive nature of contemporary social protracted conflicts. In non-Western settings, subsequently where most armed conflicts are situated, children often take on more adult responsibilities. Placed in a

14

Martin Sjøen Master Thesis University of Agder dire circumstance, joining an armed group might be the only viable option, reflecting “a determination to survive” (Hart 2008: 278). War may result in many children having to contribute in more “adult” ways within their communities. Still, it should not serve as a rationale that all children are capable of being potent active agents. Caution should generally be issued as not to exaggerate children's potential ability to be active agents, and finding a path between “agency” and “victimhood” is according to Hart and Tyrer very important (2006: 10). Singer states that two out of three child soldiers globally have had some sort of initiative in their own recruitment (2006: 61). An ILO conducted study on the use of child soldiers in central Africa reinforces this claim (2003: viii). Yet the same ILO study reveal that in hindsight, two out of three released child soldiers claim that some kind of coercion was used in their recruitment (2003: 28). What may seem like a decision based on personal initiative may in fact be indirectly coercive or manipulative. So clearly, this is a dualistic subject. A novel contribution by Brett and Specht, a book appropriately called “Young Soldiers, Why They Choose to Fight”, was in fact based on interviews with voluntarily recruited child soldiers from ten different countries (2004). Reoccurring as a red thread in this book is that the notion of “voluntary” must be placed under severe scrutiny. Schmidt also states that “voluntary” and “forced” can have different meanings in varying contexts (2007: 55). She additionally argues that using polar opposites in determining the circumstance of recruitment is not constructive, as it severely limits the spectrum of answers given in response. For some child soldiers, the recruitment process may be characterized as a mixture of both forced and voluntary factors. Forceful recruitment including recruitment by abduction has been widespread in some armed conflicts; such as the LTTEs liberation war in Sri Lanka (Kanagaratnam et al. 2005), among LRAs rebel movement in Uganda (Corbin 2008) and with the RENAMO movement in Mozambique (Boothby 2006). Some conflicts like in DRC, has seen both forced and voluntary recruitment being used (Bayer et al. 2007, ILO 2003). The circumstances of the conflict will often dictate the nature of recruitment, and this underpins the importance of generating contextually appropriate research. Furthermore, the nature of the recruitment will in many cases determine the success of the reintegration process. According to Brett and Specht, children who have not been forcefully recruited, are more likely to re-join a warring group again if the issues that lead to their recruitment is not sustainably addressed (2004: 5). To summarize the above discussion, child soldiering in some regions may predominantly be based upon coercive or voluntary recruitment. Some may have a myriad of

15

Martin Sjøen Master Thesis University of Agder both. It is likely that many former child soldiers who state that they joined an armed group on their own initiative, may have been directly or indirectly forced due to circumstances beyond their control. Share excitement has also been found, although not in great numbers, to be a reason why some children decide to join armed groups (ILO 2003: 29). In all probability, some former child soldiers who joined voluntarily, may in hindsight claim the opposite, as honest answers on such sensitive matters cannot always be expected (Honwana 2006: 14).

2.3. The impacts of child soldiering The devastating effects of violent conflict have been thoroughly documented in contemporary conflict studies (UNDP 2009). This section will therefore be limited to a momentarily exploration of the impacts of child soldiering on the young combatants themselves. Most of what has been written on the consequences of child soldiering, deals usually with its destructive outcomes (Blattman and Annan 2008a, Hart and Tyrer 2006). Leading paradigms such as Western schools of psychology and medicine has predominantly focused on the mental and physical well-being of former child soldiers. Research is prolific on deteriorating health issues and perhaps most common, the prevalence of post-traumatic stress disorder (PTSD) (Hart and Tyrer 2006: 4). These descriptions have further fueled the creation of legacies that most child soldiers are traumatized beyond repair, unable to function properly in a post-conflict setting (Blattman 2009: 231). Be that as it may, many Western paradigms have been increasingly scrutinized, as some argue they lack contextual appropriateness (Psychology Beyond Borders 2008, Wessells and Monteiro 2007). As information from a growing number of case studies on child soldiering has become available in recent years, certain tendencies are reoccurring. Many of these tendencies challenge both essentialism and the liberal perceptions that characterize most humanitarian and medical paradigms. There are certainly many former child soldiers who have experienced degenerated health issues, including but not limited to suffering from PTSD, trauma, depression, or other injuries (Wessells 2006). Psychological distress is commonly associated with former child soldiers who experience and/or perform “severe” types of violence (Apio 2008, Bayer et al. 2007, Blattman and Annan 2010). The mental well-being of former child soldiers should not be trivialized. Yet war is profoundly a social and political issue pervading all aspects of society. Paradigms that focus exclusively on “trauma” tends to medicalize and individualize the problem, thus losing sight of important dimensions of conflict (Boyden and de Berry 2007: xiv, Wessells and Monteiro 2007: 8). There are strong indications that the most common consequence of child soldiering is related first and foremost to the loss of education

16

Martin Sjøen Master Thesis University of Agder and income (Psychology Beyond Borders 2008: 2). Depending on length of time served with the armed groups, such affiliation prevents many child soldiers from completing their education, subsequently also hindering them a successful entrance to the job-market (Blattman and Annan 2010, Boothby 2006). Another decisive consequence is suggested to be related to social reintegration (Psychology Beyond Borders 2008: 24-25). The former child soldiers' immediate surroundings willingness to accept them “back” has been described as a crucial issue for their well-being (Wessells 2006). Inter-Agency Standing Committee has designed a broad framework aimed at determining appropriate intervention responses for former child soldiers (IASC 2007). It is in a form of a pyramid that divides the war-affected population into four broad layers, each layer describing which complementary support should be provided for the recipients. Approximately 10-15% of the total number of child soldiers will be placed in the top (fourth) layer. This includes those who have experienced severe atrocities, show difficulties in daily functioning, and may therefore require long term mental care follow up. The third layer contains about one third of the child soldiers, and it includes those who have experienced moderate levels of violence. They are relatively functional, yet strong community and family support would be very beneficial, and often a sufficient guarantee for their well-being. The two bottom layers include the majority. These child soldiers have experienced varying levels of violence, but show normal functionality and strong resilience. Re-establishment of basic socioeconomic services and opportunities is often the largest guarantee for their well-being (IASC 2007: 11). Some authors argue that devastating impacts of child soldiering decrease significantly over time (Annan et al. 2010). A well timed but controversial question is whether there can be any positive effects of child soldiering. Schmidt claims that children who strive to fulfil their basic needs, joining an armed group may assist them at least on a temporarily basis (2007: 62). In this context, basic needs would usually constitute providing stabile access to food and water (Wessells 2006: 23). Case studies from Liberia (Utas 2007), Sierra Leone (Peters 2004) and Uganda (Blattman 2009), reveal that some former child soldiers acquired helpful skills, became more politically engaged and expressed a strong personal growth, depending on their level and type of involvement with armed groups (Schmidt 2007: 63). Further on, post-conflict resilience which is commonly found among former child soldiers, has received more attention in recent years (Boyden and de Berry 2007: 83, Hart 2010: 1, Wessells 2006: 4,29). The dominant view in most sciences is nevertheless “rightly” that child soldiering can severely impede the lives of those involved.

17

Martin Sjøen Master Thesis University of Agder

2.4. Child soldier intervention

2.4.1. Defining DDR-(RR) The term “child soldier intervention” which has appeared in this paper, will in this section be elaborated both in terms of theoretical perspectives, and also in a more practical manner. A simplistic description would be that child soldier intervention is the process of removing underage soldiers from armed groups, and assisting them back to a civilian life. Among scholars and practitioners, the terminology “DDR”, meaning disarmament, demobilisation and reintegration is commonly used for such work (Muggah et al. 2003: 18). These efforts are part of recovery strategies encompassing “economic development, security sector reform, integration of refugees [with] justice and reconciliation” (International Peace Academy 2002: 3 in: Schroeder 2004: 3). Alternatively there is also the inclusion of rehabilitation and repatriation/resettlement as separate processes, potentially making it a DDR-(RR) process (Schroeder 2004: 30). Figure 4: Defining the DDR-(RR) process.

Disarmament: Demobilization: Reintegration: Collection and control of small Formal and controlled The process of socioeconomic arms , either voluntary or registration and discharge of support for the integration of coercive in exchange for combatants often in ex-combatants into social money or other incentives designated transit camps networks

Rehabilitation: Resettlement/Repatriation: Psychosocial and Re-location to new areas or clinical/psychiatric intervention. home country for foreign Usually carried out in the combatants reintegration process

Figure 4 explains briefly the different components that make up the DDR-(RR) process. In most cases, the process will follow the linear process of 1) disarmament, 2) demobilization, 3) reintegration, with separate components of rehabilitation and resettlement/repatriation being case dependent (Muggah et al 2003).

18

Martin Sjøen Master Thesis University of Agder

DDR-(RR) is considered to be among the most recognized post-conflict processes for disarming and reintegrating armed soldiers, and it is applied globally with conceptual differences (Muggah et al. 2003: 18). Both adult- and child soldiers are the subject of such processes, though usually through different approaches. Pauletto and Patel (2010: 42) argues that there are fundamental differences between child and adult DDR-(RR) programmes, mainly because children's involvement in conflict is a violation of their rights whether they volunteered or not. Pauletto and Patel further claim that there is a principle difference in the legal accountability between adults and children in armed conflict. Nonetheless, some post- conflict efforts do not distinguish between adult- and child soldiers, and some do not even recognize children’s needs or rights to be integrated into DDR-(RR) work altogether (Wessells 2006: 157). There are former child soldiers who have been retained from participating in intervention programmes, as they were unable to provide a hand-weapon in the disarmament process as a “proof” of their affiliation (Peter 2006). Other child soldiers may have experienced being excluded based on their ethnicity or gender (Dallaire 2011, Schroeder 2004, Tonheim 2010). Since the main emphasis in this research is to investigate the reintegration of former child soldiers as opposed to the entire DDR-(RR) process. The terminologies intervention- work, effort and program will principally be used for the remainder of this paper, unless specific parts of the DDR-(RR) process are being addressed.

2.4.2. Assessing child soldier intervention There are different conceptual views towards child soldier intervention both in theory and practice. Singer claims that intervention programmes are sustainable post-conflict tools, yet they must be approached as a process, rather than a “step”. Each component of the DDR- (RR) process must therefore be meticulously carried out adapted to the individual’s needs (Singer 2006: 183,211). Contrasting views are promoted by others who question whether intervention programmes are sufficient tools for assisting former child soldiers. In the case of Uganda, Blattman and Annan argue that the potential effect of intervention programmes is largely unknown, and that post-conflict work is more ad hoc than evidence based (2008a: 24). Generally, there is very little available knowledge on a) understanding the real impacts (positive and negative) of reintegration programmes, and b) determining whether participation in such programmes will increase the chance of successful reintegration. Some case studies reveal that child soldiers who have participated in intervention programmes are relatively pleased with their affiliation, despite various shortcomings in the respective post-conflict

19

Martin Sjøen Master Thesis University of Agder work (Boothby 2006, Humphreys and Weinstein 2004). Yet perhaps more surprisingly, Humphreys and Weinstein's survey of ex-combatants in Sierra Leone also discovered high rates of successful reintegration of child soldiers not affiliated with intervention programmes (2004: 39). Many of them even recovered faster than those child soldiers who were involved in intervention programmes. More research is certainly needed on the effects of child soldier intervention in order to determine whether such efforts can actually contribute in a positive way. According to Muggah et al., intervention programmes can be viewed on a “continuum” – from a minimalist (improving security), to a maximalist perspective (an opportunity for development and reconstruction) (2003: 5). It is reasonable to believe that the divergent approaches between a minimalist and a maximalist view, may also affect the long term success. If the goal is simply to end hostilities, then this is probably an “easier” task than addressing the causes of conflict, and potentially facilitating broad human development. It is also claimed that intervention programmes are a temporary process aimed at establishing the preconditions for development, not a permanent substitute for development. An important observation as post-conflict work of this kind should not become a surrogate for “everyday” life. Post-conflict intervention relies on the participants to eventually become self-reliant in a well-functioning way in order for it to truly be sustainable (Muggah et al. 2003). Yet according to Wessells, most intervention work to date has been aimed at improving security through the processes of ceasefire and disarmament7. A study of post-conflict intervention in Congo-Brazzaville, showed that these efforts were increasingly being approached through a “one-size-fits all framework”, a possible implication as contextual and individual consideration is argued to be a crucial factor for its success (Muggah et al. 2003: 18). Blattman and Annan also accentuate this challenge as they claim that intervention efforts in Uganda focused largely on broad-based psychosocial assistance, instead of addressing the largest impacts, namely loss of education and economic livelihood (2008a: 21). This is presumably tied to the medical dominance of Western paradigms concerning the traumatization of children in conflict. Blattman and Annan also claim that targeting child soldiers as a group in isolation can be counter-productive, “primarily because it can be stigmatizing within the community” (2008a: 21). Comparable studies have also shown that some child soldiers do not want to be singled out, as this may create hostilities or envy from other members of community who do not receive specific post-

7Wessells, M. Professor at Colombia University: Lecture in Stavanger 14.03.2012. 20

Martin Sjøen Master Thesis University of Agder conflict assistance (Boothby 2006, Wessells 2006). On that subject, ILO claims that “Reintegration programmes must target simultaneously, the youth, his/her family and the community to which he/she belongs” (2003: ix). There is now increasing acknowledgement that intervention programmes may include other vulnerable groups (so-called “non-targeted” approach), in order to gain acceptance from local communities (Blattman and Annan 2008a, Muggah et al. 2003). It seems that the debate on the success or failure of child soldier intervention is closely related to the prevailing philosophic approach (minimalist or maximalist), as well as how the practical process is carried out. To ask whether intervention work in general is successful may not be very fruitful, as any answer is surely case dependent and subjective. What the anecdotal evidence seems to suggest is that the most recognizable shortcomings of such programmes are often due to poor-timing, insufficient capacity and lack of customizing the program based on the former child soldiers’ context and gender (Blattman and Annan 2008a, Humphreys and Weinstein 2004, Peters 2006).

2.4.3. The concept of reintegration The term reintegration of ex-combatants is somehow a peculiar one. It suggests that the ex- combatants need to be supported and equipped to make their re-entry to peaceful society successful, but does not ask if there is still something into which to reintegrate. It is assumed that the physical and social structures of communities exist, and continue to function (Peters 2006: 135).

Child soldier reintegration sits at the heart of this thesis, and how it is defined influences every aspects of this process. The term itself can be problematic as it implies a “return home” for the former child soldiers, a home which may not exist anymore. In some cases, the term “integration” is probably more suitable than “reintegration”, since there are many former child soldiers who have hardly had a life outside the armed groups. Occasionally the word “reinsertion” is used instead of reintegration, which may cause considerable confusion in literature. The former is generally preferred by French based actors, while the latter is favoured by English ones according to Muggah et al. (2003: 21). Reinsertion can in some cases also refer to the share physical act of returning home, while reintegration may entail the long-term process of restoring socioeconomic networks8. The components of a reintegration process may differ according to actors and contexts. A child soldier’s background, age, served role, gender and experience will often determine the intervention

8Wessells, M. Professor at Colombia University: Lecture in Stavanger 14.03.2012. 21

Martin Sjøen Master Thesis University of Agder offered in response. In an attempt to highlight several perspectives on what constitutes “successful reintegration”, Tonheim argues that it is difficult to measure and very much case dependent (2010: 24). While the reintegration aspect may be a vague terminology, perhaps even contextually inappropriate, it is often regarded as the most important part of chid soldier intervention work (DAI 2007: 6, Humphreys and Weinstein 2004: 39-40). Coincidently, it is usually also the most difficult and long-lasting part of the intervention process (Psychology Beyond Borders 2008: 11, Wessells 2006: 160). So what has been “detached” by war and needs “reintegration”? According to Annan et al. (2010), there are two dominant perspectives in academic literature covering this. The first is that war causes injuries, loss of education, and generally lost opportunities. While the second is that war leads to social exclusion, alienation, and in the worst case psychological trauma and aggression (Annan et al. 2010: 1). Little confirmable research is available on child soldier impacts, and this statement may therefore be more assumption- than fact based. This perspective does not only presuppose these “universal” devastating impacts of armed conflict on its young participants; it also assumes that there is a gap between ex-combatants and non- combatants (Annan et al. 2010: 3). Yet very little is known on the differentiating impacts of armed conflict, or if there even is a difference between “armed” and “civilian” children in conflict. This is imperative to consider when conceptualizing the reintegration of former child soldiers.

2.4.4. Reintegration framework A conceptual framework has been created by Annan et al. (2010), which includes four major factors for assessing the reintegration process. These domains have been briefly discussed in section 2.3. It is rather speculative to claim that this is the universal reality for most former child soldiers (Annan et al. 2010: 3-7). Developed in a Western context, the actual usefulness of such a framework is still somewhat unclear. However, these four domains are put together by numerous units of analysis around the world, and they are commonly believed to be very central in the process of child soldier reintegration. Therefore will it also form the basis of this research, while being well aware of the possible implications and pitfalls.

 Social acceptance: Is related to how former child soldiers are received in their immediate surroundings including family, alternative placement, local community reintegration and host community resettlement. Social acceptance is believed to be of

22

Martin Sjøen Master Thesis University of Agder

high importance for a successful reintegration, and it is closely linked with the possibility for generating economical livelihood. Former child soldiers who are discriminated or stigmatized, will believably find it more difficult to establish themselves in a civilian setting, thus the surrounding environment will often determine the success rate of the reintegration. Embedded in this particular domain are also different sociocultural aspects such as traditional cleansing ceremonies, religious rites and other cultural customs. There are quite diverging views on whether most child soldiers are reintegrated back in a “sufficient” way. Some believe that most child soldiers generally reunite with their local surroundings in a tolerable way, while also experiencing increased acceptance over time. Others claim that former child soldiers who have perpetrated violence and female child soldiers in particular, encounter significant rejection and hostilities (Annan et al. 2010: 6).

 Hostility: Former child soldiers may be identified or perceived with having a more hostile or aggressive behaviour, due to their active participation in armed conflict. Their adaption to a non-military setting, which includes problem- and conflict solving methods in different environments, is presumed to have a decisive effect on their reintegration process. According to Annan et al., “There is much theory but little evidence on the aggression of ex-fighters” (2010: 6). Recent case studies have demonstrated that the majority of former child soldiers show modest or no aggression in post-conflict settings (Blattman and Annan 2010, Boothby 2006, Humphreys and Weinstein 2004). Perhaps this domain derives more from a constructed perception rather than a pervasive reality. It is nevertheless an important factor in the reintegration process in terms of social acceptance, family/community sensitization and support, and is therefore included into the concept of social reintegration.

 Economical livelihoods: Encompasses income earning activities and education and/or vocational training. Among the most current theories is the one that affiliation with armed groups will impact a child soldier's possibility for accumulating education, experience, wealth and human capital. Successful reintegration is believed to rely heavily on the individual’s ability to seek educational and economic opportunities. Some case studies suggest that this is perhaps the most crucial part of the reintegration process for most former child soldiers (Blattman and Annan 2008a, Boothby 2006).

23

Martin Sjøen Master Thesis University of Agder

 Psychological well-being: Psychological distress is believed to be an important step in the reintegration process, despite the increasing scrutiny based on its predominant liberal Western focus (Psychology Beyond Borders 2008). This factor is however connected with the individual experience of the former child soldier, meaning that its importance may vary greatly between former child soldiers (Bayer et al 2007). The author holds no formal education in medicine or clinical psychiatry, and can therefore not assess the levels of trauma or depression among former child soldiers. Furthermore, this domain is not directly linked with any particular objective in this research. Assessment of the psychological well-being of former child soldiers in this research relies solely on secondary data, and is merely included to provide general reference points.

2.5. A cursory overview of trends in literature Although several theories and perspectives on child soldiering have been presented in this theoretical background, a brief reappraisal of some dominant trends is still warranted. At the risk of over-generalisation, the largest contributors on literature concerning child soldiers are generally considered to come from two distinct schools of thought. Broadly summarized, these are humanitarian actors firmly rooted in (neo)-liberalism and different medical paradigms (Hart and Tyrer 2006, Gates and Reich 2010, Kostelny 2006: 19). While their highly varied and valued contributions have raised public awareness on a very important issue, they are often criticized for having a narrow focus. This has been apparent through what has already been described here as the “victimization” of children, and the almost unilateral focus on the “disastrous” impacts of child soldiering (Annan et al 2010: 1, Blattman 2009: 231). As a growing body of literature has increasingly scrutinized traditional Western contributions, appreciation for contextual sociocultural factors is now becoming more widespread (Wessells 2006: Preface). Most research on child soldiering has been interdisciplinary, focusing on post-conflict intervention efforts. Especially within social sciences, traditional supply sides of child soldiering such as poverty, war, education, culture and family are frequently highlighted (Andvig and Gates 2010: 1). The last trend to be mentioned is the apparent gap between scholars of child soldiering and the general public. Blattman and Annan acknowledge that researchers within the field of child soldiering are “often guilty of writing for a narrow audience”, thus sometimes distancing themselves from mainstream media, policymakers and people in general (2008b: 22). While there is an ever growing body of academic literature on child soldiering, it still finds itself

24

Martin Sjøen Master Thesis University of Agder firmly placed behind other more non-scholarly or “popular” science paradigms.

2.5.1. Notable research gaps Thorough work on research gaps in the field of child soldiering has already been made available by Psychology Beyond Borders (2008) and Tonheim (2010), yet here are some key facets. Existing research on child soldiering is often categorized as geographically “scattered”. Andvig argues that satisfactory data, “not normative presupposition” on child soldiering is available from a limited number of places. These are first and foremost Sierra Leone and Sri Lanka, and could possibly be developed for Liberia and North Uganda (Andvig 2006: 10). According to Tonheim, additional research from the African continent includes smaller case studies from Angola and Mozambique (2010: 26). Longitudinal studies are almost absent in the African region all together, besides the study of RENAMO soldiers in Mozambique (Boothby 2006). In correspondence with Andvig over available research on child soldiers in DRC, he stated that literature here is “interesting but not very scientifically oriented”9. Larger research projects in the Great Lakes region in Africa include the International Labour Office’s survey on child soldiering in Rwanda, Burundi, DRC and Congo-Brazzaville (ILO 2003), and the medical study on the association of PTSD and trauma levels among former Ugandan and Congolese child soldiers (Bayer et al. 2007). The Norwegian research institute Centre for Intercultural Communication (SIK) has published one report on the topic, and they are currently working on different research on child soldiering including a longitudinal study (SIK 2012). A handful other studies have also been conducted on post-conflict challenges for former child soldiers in DRC, such as an assessment of disarmament and reinsertion in the Ituri district (Bouta 2005), evaluation of multi-country DDR in the Great Lakes region (DAI 2007), a policy dialogue to advance child soldier demobilization (Agborsangaya 2000), and DDR challenges in Eastern DRC (Pauletto and Patel 2010). Yet empirical evidence from the contextual setting of DRC is nearly non-existent. Further information on child soldiering in this region has been conducted by humanitarian organizations, for example Amnesty International (2003), Coalition to Stop the Use of Child Soldiers (2010), Save the Children UK (Verhey 2003) and UNICEF (2003). Gender perspectives including the circumstances of female child soldiers suffers from vast research gaps in general (Psychology Beyond Borders 2008, Tonheim 2010). To the author's limited knowledge, there is currently only one exclusive female study in the context

9 Andvig, J. A. Senior Researcher at the Norwegian Institute of International Affairs (NUPI). E-Mail: 12.01.2011. 25

Martin Sjøen Master Thesis University of Agder of DRC that includes interviews with female child soldiers. Tonheim, a research fellow at SIK investigated levels of stigmatization for girls formerly associated with armed groups in Eastern DRC (2011). A conceptual study on the incorporation of gender perspectives into Congolese DDR processes has further been conducted by Schroeder (2007).

2.6. Contextual background

2.6.1. Brief political history

Figure 5: Map of DRC. Source: World Gazetteer (2012). In Congolese traditional settings, historic recounts have customarily been passed down verbally from generation to generation. Written testimonies were not introduced until the Europeans arrived in the fifteenth century (Boya 2010: 20). Conversely, the authentication of ancient Congolese history including political systems, culture and tradition, derives mainly from the initial Western explores own perceptions and interpretations. The region of modern day Congo was inhabited by Bantu speaking groups who formed various kingdoms in the central African region between the tenth and fourteenth

26

Martin Sjøen Master Thesis University of Agder century (Britannica 2011). Initial contact with the external world occurred in 1482, when Portuguese explorers navigated through the Congo River. Trade routes developed between the Europeans, Arabs and native Congolese, in due course influencing social identities in Congo (McCalpin 2002). During the Berlin Conference in 1984-85, Africa was divided between colonial powers and Congo was declared personal property of the Belgian King Leopold II (McCalpin 2002: 35). He renamed it “Congo Free State”, and Location: Central Africa systematic exploitation of Congolese people and Capital: Kinshasa Total area: 2,345,410 sq km extraction of natural resources was widespread Population: 71,712,867 during his rule. Between 1885-1908, half of Life expectancty at birth: 55.3 years Median age: 17.4 years Congo's population perished due to famines and Poverty rate: 71% of population state-sanctioned violence, while artificially Language: French, Lingala, Kingwana, Kikongo and Tshiluba created provinces by the administration Source: Britannica (2011). exacerbated existing ethnic tensions (McCalpin 2002: 35). After international criticism and pressure, Congo was handed over to Belgium, becoming its official colony in 1908. Human fatalities decreased under the more “effective” Belgian rule, yet the exclusion and discrimination of Congolese natives was nevertheless consistent (McCalpin 2002: 36). Ubiquitous human and mineral exploitation continued all through Belgium’s rule of the Congo. Pro-independence movements gathered momentum after the Second World War, and in June 1960, Belgium withdrew abruptly from Congo. Following the independence, the country changed its name to the “Democratic Republic of Congo” in 1964. Social and political turmoil was omnipresent in Congo DRC during its first years of independence. This led to Colonel Joseph Desiré Mobutu, later President Mobutu Sese Seko, seizing power in a coup d'état in November 1965 (McCalpin 2002: 40). In 1971, he renamed the country “Zaire” while gradually moving it towards autocratic rule. Poor economic development, ethnic tension and political uncertainty coupled with regional instability worsened the development in Zaire (McCalpin 2002: 41-42). By early 1990s, the country had not experienced any political or economic development for more than twenty-five years, and the excessive government spending resulted in hyperinflation. This crisis was central in the ultimate collapse of Mobutu's rule as the weakened regime experienced eroding control in periphery regions of Zaire (McCalpin 2002: 45). General elections were announced several times, but canceled to public frustration. Yet it would be the 1994 genocide in neighboring state Rwanda that would eventually led to Mobutu's downfall, subsequently having a decisive effect on Zaire for years

27

Martin Sjøen Master Thesis University of Agder to come (McCalpin 2002: 46).

2.6.2. Historicity of the Congolese Wars Africa, Franz Fanon famously remarked, has the shape of a pistol, with Congo-Kinshasa resembling the trigger housing … the pistol, however, rather than pointing towards Antarctica, aims its fire inwards (Young 2002: 13).

The social protracted violent conflict in DRC is a multifaceted one where politics, resources and ethnicity are all interchangeably linked. Fundamental views differ in explaining the violent conflict in DRC (Clark 2002: 2). Yet similar to the previous section, the following can only offer a very brief and subsidiary overview of the contemporary warfare in DRC. Although several perspectives that are presented may be viewed as contested, there will be no attempt to provide systematical analysis of key historic events, the actors involved or their motives. Undoubtedly a weakness as the “crisis in the Congo cannot be understood as an event in isolation” (McCalpin 2002: 33).

2.6.3. First Congolese War (1996-1997) As the Rwandan genocide unfolded in 1994, it left more than 800,000 human beings murdered within the space of one-hundred days (Dunn 2002: 55). This triggered an outpouring of between one and two million Rwandan refugees settling in Zairian refugee camps. Many fugitives were former Interahamwe and Rwandan army soldiers (FAR), who predominantly had carried out the Rwandan genocide (Murison 2002: 225, Orogun 2002: 29). Re-organized in Zairian refugee camps alongside fleeing civilian Rwandese, many soldiers launched attacks on the new Rwandan army (RPF). Violence was also carried out towards the civilian population in Zaire (Dunn 2002: 55). These “génocidaires” would later form the highly notorious armed group FDLR. At the same time, the Ugandan armed groups Allied Democratic Forces (ADF) and Lord's Resistance Army (LRA) launched attacks on Uganda from their respective locations in northern Zaire (Reyntjens 1999: 242). Failure or reluctance by Mobutu's central government to act upon these violent outbursts in eastern DRC, were followed by increased ethnic tension between the ethnic Congolese Tutsis (Banyamulenge), Zairian population and Hutu refugees. According to Dunn, this resulted in the Rwandan government and the Banyamulenge deciding to take matters in their own hands, by “attacking their attackers” (2002: 56). In October 1996, all Banyamulenge living in South Kivu were told by the regional government that they had to leave the country within one week (Dunn 2002: 56). This act was central in sparking an uprising which eventually led to the First Congolese War. The Zairian 28

Martin Sjøen Master Thesis University of Agder national army FAZ, Interahamwe and refugee camps were directly targeted in violent attacks by a loosely formed alliance of armed groups. Led by a relative unknown “career rebel” named Laurent-Desire Kabila, and heavily backed and orchestrated by Rwanda and Uganda, the alliance formed the ADFL movement (Dunn 2002: 56). They were soon to be joined by troops from Angola, Burundi and Eritrea in an attempt to overthrow the Zairian president, while the Angolan armed group UNITA was to partake on Mobutu's side (Dunn 2002: 57). Referred to by many Congolese as the “War of Liberation” (Agborsangaya 2000: 3), this was the onset of the “World War of Africa” (Tonheim 2011: 3). By May 1997, much of Zaire was in control by ADFL including the capital Kinshasa. Mobutu in due course fled the country and L. D. Kabila proclaimed himself president while renaming the country the “Democratic Republic of Congo” (Orogun 2002: 25).

2.6.4. Second Congolese War (1998-2003) Shortly after he was placed in power, L. D. Kabila began to distance himself from his supposedly closest allies Rwanda and Uganda, partly due to strong internal pressure (Reyntjens 1999: 245). The Congolese president commanded all foreign troops to leave the country immediately, a maneuver which provoked his “key external backers” (Dunn 2002: 62). Comparable to that of his predecessor Mobutu, L. D. Kabila did not act upon the ethnic challenges, nor did he address the violent attacks being carried out in and from DRC (Dunn 2002: 62 and Orogun 2002: 30). Rwanda and Uganda soon found themselves facilitating yet another rebel attack on its neighboring country, this time through the armed groups RCD and MLC (Orogun 2002: 26). By August 1998, this rebellion proved quite successful gaining territory at a rapid speed. Yet L. D. Kabila was able to call on military aid from other African nations through the Southern Africa Development Community (SADC) framework (Reyntjens 1999: 247). Within weeks, the likes of Angola, Zimbabwe and Namibia came with their support, fighting alongside the new Congolese army FAC (Dunn 2002: 63). Rwanda and Uganda would eventually be accompanied by Burundi, while Chad, Sudan and Libya provided both direct and indirect assistance to L. D. Kabila (Reyntjens 1999: 247). On top of this, several non-state militias and armed groups were also involved on both sides of this war (Orogun 2002: 31). Congolese FAC soldiers would frequently give their support to groups whom they allegedly were at war against. At the same time, Mayi-Mayi, a collective term for local Congolese militia groups who previously had fought against Kabila, now aligned with him (Coalition to Stop the Use of Child Soldiers 2010: 2, Reyntjens 1999: 243,248). The Second Congolese War caused great human suffering, frustration and was indeed

29

Martin Sjøen Master Thesis University of Agder very expensive for the countries involved. This resulted in the signing of the Lusaka Peace Accord on 10 July, 1999. (Dunn 2002: 67). Despite several of the involved countries and armed groups signing this accord, the fighting was in effect consistent. Dunn relates this to financial incentives based on mineral extraction, claiming that DRC had become an “economic colony for most combatants, with its resources being drained to the east and south” (2002: 70). A seemingly military stalemate characterized the conflict, and in 2001, L. D. Kabila was assassinated by one of his own bodyguards, an act still surrounded by ambiguity (Dunn 2002: 68). His son Joseph Kabila was sworn in as successor in 2001 (Dunn 2002: 69). The new president “resuscitated” the Lusaka Peace Accord in 2001, allowing international observers from the UN operation in DRC (MONUC) to enter the country (Dunn 2002: 69). A series of UN and South African sponsored talks achieved a breakthrough when a historic transitional government was formed in 2002. At the same time, an inter-Congolese dialogue between warring parties provided temporary success (Dagne 2011: 3). This culminated in the formalized agreement to integrate parts of armed groups such as FAC, MLC, RCD and Mayi- Mayi into a new Congolese regular army called FARDC in 2002, and end all hostilities (Baaz and Stern 2008: 63). What had presumably been the deadliest conflict since the Second World War with debatable estimates of nearly four million deaths since 1998 (Clément 2009: 91, Coghlan et al 2006: 1), was officially ended in June, 2003 (Autesserre 2007: 423).

2.6.5. Sub-conflicts of the Congolese Wars (2003-Present) This section deals with the protracted violent conflict(s) that has occurred mainly in northeastern DRC, since the official end of the Second Congolese War. Sometimes referred to as the Ituri Conflict and Kivu Conflict(s) respectively, for simplicity, these dispersed fighting’s will be referred to simply as sub-conflicts of the Congolese Wars. The apparent success of the Sun City peace agreement signed in South Africa had been limited as periodic fighting reoccurred (Tonheim 2011: 3). Factors like ethnicity, minerals, citizenship, security and grievance were still fueling the violent conflict resulting in frequent massacres, mass displacement and human-rights violations. More than a year after the peace agreement had been signed, a “thousand civilians continued to die every day” (Autesserre 2007: 423). Regular Congolese soldiers (FARDC) and armed groups operating within this region, were consistently perpetrating violent acts for the control of natural resources mainly in the eastern parts of DRC (Autesserre 2007: 431). At the same time, the UN force MONUC (renamed MONUSCO in 2010) were criticized for not providing sufficient protection to the civilian population, and even committing human-rights abuse themselves (Dagne 2011: 9).

30

Martin Sjøen Master Thesis University of Agder

While most foreign troops had officially withdrawn from the country, evidence suggests that neighboring countries and Rwanda in particular, were still very much active in the violent conflict and mineral exploitation (Autesserre 2007: 427). Strong allegations were especially aimed at the Rwandan support for RCD-Goma, remainders of the fractioned military group RCD, as well as the armed Tutsi group CNDP, spearheaded by General Laurent Nkunda (Thakur 2007: 57,59). The constant threat of armed groups for both DRC and its neighboring countries, led to a series of futile military operations in collaboration between DRC, Rwanda and UN respectively (Dagne 2011: 8-9). While many armed groups have seen a decline of personnel in their ranks, violent attacks continue to occur (Security Council 2010). The general situation in DRC is seemingly more stable now than it was during the latter stages of last decade. Yet, there are still numerous armed groups actively engaged with mineral extraction in the northeastern region. Human-rights abuses are still present, and the many civilians living in rural regions are occasionally targeted by the myriad of armed groups that operate here.

2.6.6. Child soldiers in DRC As previously discussed, a pervasive development in armed conflict has been the deployment of children and adolescents, often in large numbers as active child soldiers in combat (Singer 2006: 4-5, Wessells 2006: 3). During the armed conflict, between 30,000 – 50,000 child soldiers or “Kadagos” (little ones), have been active in DRC (Bell 2006: 4, Singer 2006: 21, Wessells 2006: 13). Although the figure of active child soldiers in DRC has been reduced somewhat the last few years, there are still at least 7-8000 child soldiers remaining, mainly within the northeastern parts of the country (Pauletto and Patel 2010: 36). Coalition to Stop the Use of Child Soldiers estimates that nearly 40,000 child soldiers have been released from armed groups in recent years (2011: 10). Quantifiable statistics are however scarce in this region and these figures are perhaps more the result of educated guesses than accurate science. Yet judging from the number of child soldiers who have passed through intervention programmes10, the higher estimates are probably more authentic. All armed groups both national and foreign, have recruited child soldiers in various degrees. In some armed groups, underage soldiers amount for as much as one-third of the total number of soldiers according to the UN (Achvarina 2008: 2). Girls are believed to make up

10At the start of 2007, 54,000 children had according to Thakur passed through official intervention programmes, several thousand had been repatriated to neighbouring countries, and an estimated 15-20,000 were still awaiting formal assistance (2007: 54). At least 11.000 had self-demobilized according to UNICEF (Coalition to Stop the Use of Child Soldiers 2008: 4). However, it is imperative to consider that both under- and over- reporting may implicate these figures. 31

Martin Sjøen Master Thesis University of Agder approximately 40% of the total number of child soldiers in DRC, serving a wide arrange of roles just like their male counterpart (Bell 2006: 4). Still, girls represent only a minority of the total number of child soldiers who had passed through intervention programmes (Coalition to Stop the Use of Child Soldiers 2008: 4). Findings by Tonheim suggest that female child soldiers experience significant levels of stigmatization upon reintegration, potentially “hampering” this process (2011). Most child soldiers are recruited in their teens between the ages of twelve and eighteen, and both forced and voluntary recruitment has been present (Bayer et al. 2007: 557, ILO 2003: 37). The lack of verifiable knowledge impedes any attempt to conclude which approach has been the dominant one. Although some research suggests that the majority of child soldiers had some initiative in their own recruitment, many respondents cited lack of security, food and widespread fear as important reasons for joining armed groups (ILO 2003: 29). Recruitment factors for many child soldiers may therefore be a mixture between voluntary and indirect coercive. Abduction-led recruitment has also been frequently used. Re- recruitment figures are speculated to as high as 60% in eastern parts of the country (ILO 2003: 37). Child soldiers in DRC generally express a high level of both witnessing and participating in severe violent acts (Bayer et al. 2007: 558, ILO 2003: 43). The considerable need for child soldiers in DRC corresponds closely with the almost uninterrupted violent conflict that has characterized the country since the mid-1990s. This warfare has every trademark of contemporary protracted social conflict including prolonged fighting, relative little direct combat between belligerents, large flows of displaced population and direct targeting of civilians (Gates and Reich 2010: 8). The armed groups’ demand for cost-effective combatants combined with child labour traditions are likely considerable factors in explaining the vast use of underage soldiers. People aged fifteen and younger constitute about half the population in DRC, reinforcing the great availability for potential child soldiers (Britannica 2012). Availability alone should however not be viewed as a causal explanation for why child soldiering is so widespread here. Formalized child soldier intervention in DRC dates back to the period between the two Congolese Wars, and numerous international NGOs and national CBOs conduct work of this kind throughout the country (DAI 2007: 30). In addition, DRC has also collaborated in a multi-country demobilization and reintegration program (MDRP) with neighboring countries. Still, perhaps as little as 50% of returning child soldiers have had access to “reintegration support” (Coalition to Stop the Use of Child Soldiers 2011: 2). There is also notable difference between child soldier intervention work in DRC, and comparable work in Angola,

32

Martin Sjøen Master Thesis University of Agder

Liberia and Sierra Leone. DRC is to some extent still characterized by armed conflict, as opposed to the other African countries, where intervention work occurred mainly in post- conflict environments. Therefore, child soldier intervention efforts as traditionally defined may not be entirely appropriate in DRC according to Pauletto and Patel (2010: 48). International pressure and increased focus on the illegality of using underage soldiers was an important step into reducing the amount of child soldiers in DRC. Following the attempt to unify several armed groups into a national Congolese army in 2002, the criminalization of recruiting child soldiers became increasingly more important. This had some profound effects. Thousands of child soldiers were released from their respective armed groups, many sent to join formalized intervention programmes, others to self-demobilize. Armed groups reluctant to comply with the army unification adopted approaches to become even more covert in their child soldier usage and recruitment, sometimes even abandoning child soldiers to care for themselves. Some groups have disregarded all developments made while continuously recruited underage accomplices (Coalition to Stop the Use of Child Soldiers 2008: 2). In March 2006, the Congolese Major Jean-Pierre Biyoyo became the first person to be convicted for recruiting child soldiers, a case which established an important precedent in international law (Chikuhwa 2010: 48). Another case from DRC, rebel commander Thomas Lubanga was the first militia leader brought to the International Criminal Court (ICC) for child soldier recruitment (Beber and Blattman 2011: 1). At the time of writing, the international criminal court has delivered the “first verdict in its 10-year history”, finding Thomas Lubanga guilty of recruiting child soldiers (The Guardian 2012).

2.6.7. Choice of sites The cities of Goma and Bukavu, each constituting the capital of North- and South Kivu respectively, have been selected for the purpose of data collection. In addition, several smaller townships and villages in vicinity of these cities have also been a part of the research process. This geographical and subsequently methodological limitation rests on issues related to personal security and practicality. Much of northeastern DRC is still characterized as unsafe, with consistent occurrence of violence, human-rights abuse and pillaging (Eastern Congo Initiative 2011). Roadside travel between larger cities is described as hazardous, and due to the solitary nature of this research, a choice was made to prioritize the safety of the author and those involved. A certified strength with the choice of these locations however, is firstly that the contemporary conflict in DRC has been deeply embedded in this region (Dagne 2011: Summary). Goma and Bukavu have been at the very centre of the violent conflict in

33

Martin Sjøen Master Thesis University of Agder

DRC since the outbreak in the 1990s, and child soldiering has been rampant here (ILO 2003: 17). Secondly, child soldier intervention work in DRC stems largely from the Kivus (DAI 2007: 30). Thus a great potential lies in gathering reliable data on the long term effects of child soldiering including reintegration aspects from these sites. Despite that, satisfactory data on demographic trends and socioeconomic figures from Bukavu and Goma is generally scarce. This section is therefore based on various broad- ranged reports from North- and South Kivu, some more up to date than others. For that reason, the following segment is best viewed as nonspecific background information rather than precise data. Bukavu: Located at the extreme south of Lake Kivu sits Bukavu with its faded reminiscence of a once great colonial leisure site. Yet very little scenic beauty has endured of the town itself, as uninterrupted totalitarian rule and armed conflict has eroded this lakeside city. As the capital of South Kivu, Bukavu was once regarded the main political and economic center in the Kivus. The population is calculated to be approximately 806.940 (World-Gazetteer 2012). Comparable to the country overall (African Development Fund 2004: 1), agriculture is the single most

Figure 6: Map of study sites. important economic sector for the Source: PG Kivu (2012). population in Bukavu (Eastern Congo Initiative 2011: 304). While mining exports are believed to account for the largest part of state revenues, farming is still the principal sector in terms of GDP and employment. In addition, the informal sector such as product trading is considered to be notable, yet this is not reflected in official GDP or export valued statistics (CIA 2011). Agriculture and other primary sectors have experienced a significant decline during the subsequent wars in DRC, resulting in a situation with “extremely worrying impoverishment of the populations and food

34

Martin Sjøen Master Thesis University of Agder insecurity” (African Development Fund 2004: 1). Income surveys in northeastern DRC reveal that more than 90% live on less than US$2 per day, while some 85% live on income below US$1 per day (Vinck et al. 2008: 22). Healthcare and educational sector is more or less in abysmal state, sometimes nonexistent. With a public sector unable to provide sufficient education and healthcare, different NGOs, CBOs and religious institutions are often the largest providers of these services. The highly centralized Congolese state has equally been incapable of improving regional infrastructure (Trefon 2011: 49). Besides scattered road construction efforts made by international donors, most roads and railway systems have not been properly maintained since the colonial powers left in 1960. Bukavu has in recent years attracted scores of people to its vicinity. Research shows that nearly one in seven residents in South Kivu was displaced as of late 2010, reinforcing the already existing poverty and food shortage (Eastern Congo Initiative 2011: 299). The Eastern Congo Initiative estimates that the unemployment rate in Bukavu is 28.3% (2011: 304). Yet rural to urban migration in Bukavu persists, as it represents safety for civilians trapped in conflict prone rural regions. Development of (il)legit mining especially the coltan “boom”, has seen a rapid growth of small yet very rich elite (Jackson 2003: 15). Urban slums, luxury hotels, banks, financial houses and shopping malls are simultaneously expanding at a very rapid speed. Goma: Situated at the northern opposite of Lake Kivu sits Goma. Both Kivu provinces are part of the Western Rift Valley, and it contains DRCs highest and most rugged chains of mountains (Britannica 2011). Goma is overlooked by the highly active volcano Nyiragongo. With its last major eruption in 2002, much of the city centre was ruined, leaving more than 120.000 people homeless and 300.000 - 400.000 refugees (Allard et al. 2002: 2). Goma shares many socioeconomic features with Bukavu, yet it has probably passed the latter in terms of regional political and economic importance (Eastern Congo Initiative 2011: 146). North Kivu and Goma in particular, is the only region in DRC with any notable tourism. Population was in 2008 estimated to be around 550.000, but has surely increased since then (UNDP – DRC 2008: 34). Similar to Bukavu, most employment is in agricultural sector followed by fishing, yet commercial trade and mining has become more illustrious. Agriculture and other divisions such as cattle ranching have suffered greatly from the violent conflict, with the consequence of widespread poverty and extensive malnutrition. Education and healthcare services are limited, with international organizations often being key beneficiaries. Equivalent to South Kivu, women are generally the most marginalized group in

35

Martin Sjøen Master Thesis University of Agder terms of education and employment (Eastern Congo Initiative 2011: 144). The initial rebellion to “oust” the First Congolese War emerged in Goma, and enormous population growth has characterized the city ever since. Goma has comparable to Bukavu, also become an attraction point for displaced civilians, many seeking refuge from an uncertain rural livelihood (Jackson 2003: 16). Approximately 81% of the population has experienced being displaced since 1993, denoting the volatile conditions in Kivu region (Vinck et al. 2008: 29).

2.7. Theoretical framework Existing knowledge, or theory, serves as the foundation of this research (Khan 1999: 3-4). A preliminary theoretical framework contributes to problematizing the issue under investigation and identifies the different variables that will be measured in the work. Furthermore, it helps to clarify the researcher’s objectives and values. Embedded within the interdisciplinary field of social science, this research draws on already discussed theories and concepts situated within the field of child soldiering. It utilized also concepts from contemporary conflict management within peace and conflict studies. Most prominent here is the theory of conflict transformation. The philosophy of conflict transformation may differ between actors and contexts, and this thesis make use of important concepts from the founder of modern peace studies Johan Galtung (1996), which were later revisited by Lederach (1997), Miall (2004) and Ramsbotham et al. (2008). Within the field of peace studies, conflict transformation is sometimes regarded as an extension of conflict resolution and conflict management respectively. A noteworthy difference between the three philosophies however, is that conflict transformation seeks not to manage or reduce hostilities per se, its main goal is rather to address the root-causes and structures which sustains the conflict (Ramsbotham et al. 2008: 14, 29). This is often through the transforming of “relationships, interests, discourses” and even the “very constitution of society that supports the continuation of violent conflict” (Miall 2004: 3-4). Reduction of hostilities is a natural effect of this transformation. In this particular setting, conflict transformation can potentially be regarded as the “deepest” level of peacebuilding approaches. Another concept which will be briefly utilized in the theoretical framework is the paradigm of human security. Through this paradigm, the perspectives of community members will be included to provide the aspects of additional stakeholders. Developed in the post-Cold War setting, human security was first articulated internationally in the 1994 UNDP Human

36

Martin Sjøen Master Thesis University of Agder

Development Report (Jolly and Ray 2006: 4). This holistic security paradigm challenges the traditional state-centric perception of national security, as it views all threats against the “individual” to be more important than external aggression towards the state. Briefly described, it involves several broad issues of security including protection from war, poverty and disease (Jolly and Ray 2006). Gradually, human security has been divided into two interlinked fractions, or the “freedom from fear” approach (protection from war), and the “freedom from want” approach (protection from hunger and disease) respectively (King and Murray 2002). Within this specific research, the main emphasis will be the “freedom from fear” approach, as the focus is on analyzing the consequences of armed conflict for the individuals and the communities (Gates and Reich 2010: 247). Demilitarization of armed groups and the reintegration of military personnel into civil society is a central element of this approach (Jolly and Ray 2006: 4). Comparable to the field of child soldiering, peace and conflict studies is not exempt from criticism. Western approaches to peace studies are repeatedly criticized for not being universally applicable in non-Western settings (Ramsbotham et al. 2008: 7). Some also argue that conflict transformation and human security are vague philosophies loosely assembled from different theories, with severe practical implications. Still, as both these concepts are based on a conflict-in-context approach with extensive consideration of crucial background aspects, they are argued to be applicable in protracted contemporary conflicts (Miall 2004).

2.7.1. Applying the theoretical framework The practical implementation of the theoretical framework in this dissertation is as following: Part one and two of the analysis presents demographic features and common socioeconomic traits of former child soldiers when reintegrating back to society. Hence these two parts are investigated through the conceptual framework as defined by Annan et al. (2010), first described in section 2.4.4. This part looks at the general reintegration process for former child soldiers considered through leading theories on child soldiering, thus it is not analyzed through distinct theories of conflict transformation. Part three looks at the effects of formalized intervention programmes. Typically these programmes presuppose that successful reintegration meets the socioeconomic reintegration needs of the individual child soldier, ultimately avoiding re-recruitment, and assisting the participants on the path to become well-functioning and productive members of their respective communities. These intervention efforts will be analyzed mainly through theories

37

Martin Sjøen Master Thesis University of Agder of conflict transformation. Still, child soldier reintegration cannot be viewed in isolation, despite it being the main emphasis in this thesis. The needs of local communities such as their security, acceptance, and recognition are all prerequisites for a successful intervention process. Security is by large a precondition for human development (McIntyre and Weiss 2003). Conversely, the conceptual framework of child soldier reintegration and transformational impact of intervention efforts should be included in a larger framework, encompassing elements from conflict transformation theories and the human security paradigm respectively.

Figure 7: Theoretical frameworks model.

Figure 7 assumes that successful reintegration of former child soldiers is conditional on 1) the cause of child soldier recruitment is addressed, 2) the socioeconomic reintegration needs of former child soldiers are met, 3) interpersonal relationships between former child soldiers and local communities are transformed in a positive and constructive way, and 4) broad human security is provided to the mentioned actors.

38

Martin Sjøen Master Thesis University of Agder

Chapter 3: Methodology 3.1. Research strategy Scholarly methodology involves a systematic research procedure that is used for studying a particular phenomenon. The methodological approach is based on a research strategy that serves as a framework for the data collection and subsequent analysis. Traditionally, research strategies have been divided into two fractions: qualitative and quantitative. A simple distinction between the two is that the former extract empirical data from the “written or spoken word”, while the latter emphasizes quantifiable statistics, or “numbers”. The choice of a research strategy is influenced by several factors, including the researcher’s epistemological views, or the philosophical study of knowledge, and ontological views, or the philosophical study of reality (Bryman 2008). Another issue of importance is the researcher’s stance on data interpretation including the analytical approach. This research is placed within the qualitative division. Among others, this approach involves more in debt investigation, less structured research design and a recognizable limitation in the apparent loss of external validity (Bryman 2008). Moreover, it holds an interpretive epistemological view which rejects universal truths (anti-positivism). In so it argues that the social realm is contextually defined, and should therefore not be studied in the same manner as the “natural world”. That implies to issue general caution around predetermined Western ideas on the phenomenon of child soldiering including the reintegration process. It is not sufficient to describe social phenomenons superficially; they must also be “understood”. Ontologically, the research lends itself towards social constructionism, meaning that reality is perceived as constructed in a dynamic process by people within the given context (Bryman 2008: 19).

3.1.1. Research design The chosen design in this research is case study which entails vigorous analysis of a given phenomenon or group of people within a context (Bryman 2008: 52). In this setting, the case or “object” is the investigation of child soldier reintegration in the Kivus, and specific socioeconomic factors are the measured variables. Case study should be a fitting design since the object of the research is comprehensive investigation within a geographical limited region. Conflicts where child soldiering is present tend namely to be localized within communities, regions and ethnic groups (Ames 2010: 16). This research incorporate features from ethnographic study as social and cultural domains are explored. Given the fundamental disposition surrounding child soldiering, including scarcity of previous research and

39

Martin Sjøen Master Thesis University of Agder difficulties in obtaining primary data, this research is exploratory of nature. Exploratory research is often characterized by not being clearly problematized, and it is customarily difficult to generalize upon its findings. Multiple methods have been used in this research as it may reinforce the credibility of the empirical findings. Credibility can be understood as a sort of equivalent to reliability and validity of quantitative methods. A multitude of applied methods is commonly referred to as method triangulation (Halvorsen 2009: 298). In using different methods together, the researcher might find that the result of information gathered, equals more than the sum of the methods (Bryman 2008: 379). The methods that account for this triangulation are in debt interviews, focus groups and participant observation.

3.1.2. Participatory research A key factor in this research is the participatory use of former child soldiers, a notable gap both in DRC and the field altogether. Participatory research is here defined as generating empirical knowledge through the “merging” of local and academic knowledge11. With that said, it is important to distinguish between the roles of participating respondent and researcher in terms of scientific value. Rye for example, argues that participatory research is characterized by some noteworthy challenges, potentially rendering it a controversial research tool12. Although participant perspectives will be integral in this work, respondents' views will also be the subject of an individual interpretation by the researcher. The scientific contribution is therefore exclusively the researchers, albeit it being interpreted from the participants’ spoken word. Several considerations should also be made when the objects of investigation are human beings in sensitive settings, including the premises of how the research is conducted. There is however a belief among some researchers that active participation in ethically sound research can spark personal benefits for those involved (Hart and Tyrer 2006: 4-7). In that sense, it entails a feature which may be comparable to that of action research, namely that the research should ideally be useful for the participating respondents. But unlike action research where the aim is to “work on the diagnosis to and solution of a problem” in collaboration with members of a social setting (Bryman 2008: 382), participatory research should preferably aspire to reward the respondents on a more personal level. This may include strong personal growth, self-empowerment and critical reflections for the respondents (Hart 2010). Participatory research may on the other hand also risk turning harmful for some respondents,

11McKay, S. Professor at University of Wyoming. Lecture in Stavanger 14.03.2012. 12 Rye, S. A. Associate Professor at University of Agder. E-mail 21.03.2011. 40

Martin Sjøen Master Thesis University of Agder as there is an ever present chance of evoking devastating memories or even jeopardizing one’s personal safety. Furthermore, active “participation” is generally an acknowledged principle in sustainable human development (Hart and Tyrer 2006: 13).

3.2. Sampling Due to the sensitive nature of the topic, sampling of former child soldiers proved a substantial challenge, especially since their identities were not easily accessible. Non- probability purposive sampling were respondents are selected based on their assumed thematic appropriateness, was therefore utilized in this work (Bryman 2008: 458). While the objective of this study is former child soldiers, other stakeholders such as local communities, professionals and field-experts were also included in the investigative process. For practicality, a methodological decision was made to focus exclusively on male former child soldiers, preferably those eighteen years or older who had been demobilized and reintegrated a year ago or more. Respondents are therefore considered mature by practical application, and adults as defined by “international law”. To exclude females all together is certainly a limitation, but to comply within the boundaries of this research, a compromise was made. Moreover, with the author being male, the appropriateness of interviewing female child soldiers, many of which have experienced ghastly sexual- and human abuse by men was also considered. The characterization of these adult respondents as former “child” soldiers may be disputed, as there is a clear divide between children and adults. Yet, the categorization as child soldier is accompanied with a set of indisputable rights, at least theoretically, which are important to oblige to. Thus this characterization is rightfully maintained throughout this thesis. Initially, collaborating organizations would suggest former child soldiers who had passed through intervention programmes as potential interviewees. This type of sampling is typically referred to as convenience or opportunity sampling. Such non-probability sampling is characterized by being somewhat easily applicable, and useful especially in the opening stages of the research (Bryman 2008: 183-184). Respondents would later identify other prospective interviewees, subsequently incorporating elements of snowball sampling into the research. Through this approach, the author was able to interview a small number of former child soldiers who had not passed through intervention programmes (non-affiliated). The relatively low number of non-affiliated respondents was due to difficulties in tracing and verifying actual child soldiers in local communities. Both convenience and snowball sampling is commonly used in qualitative studies. Advantages with these forms of sampling include

41

Martin Sjøen Master Thesis University of Agder less preparation and they are easier to administer. Such strategies are perhaps more compatible with case studies, but it will also reinforce the challenge of representativeness and external validity (Bryman 2008: 185). What is more is the scientific dilemma of having the majority of respondents selected beforehand by organizations, whom themselves have discrete or indefinite agendas. The circumstance of this research was however dependent on collaboration with external organizations, and selection criteria was further used in an attempt to reduce bias. Accordingly, Ames argues that NGOs often have the required skills and resources to carefully create selected units of observations within conflict regions. Loss of generalization could therefore be compensated by a gain in the accuracy of the “theoretical argument” (Ames 2010: 23). An effort was also made to conduct interviews and/or focus groups with members of the general population, within selected communities. Unfortunately this work was not able to conduct separate interviews with parents or other relatives of former child soldiers. Prior to the fieldwork, extensive effort was made to establish contact with several national and international organizations devoted to child soldier intervention work. Those willingly were interviewed before, during and after the fieldwork. Some interviewees suggested other professionals and field-expert who could be relevant for this research, thus sampling here was a combination of purposive and theoretical sampling. The latter involves the distinguishing of respondents based on their theoretical relevance (Bryman 2008: 414).

3.3. Data collection Primary data was collected in the months of January and February 2012. The investigative process was first established in Bukavu and vicinity located in South Kivu. Pilot testing commenced shortly after arrival, and all interviews with former child soldiers were conducted in this region. After Bukavu, the investigative work moved northwards to the city of Goma, situated in North Kivu.

3.3.1. Interviews The vast amount of data was collected through in debt semi-structured interviews in this research. Semi-structured interviews are less formalized and rigorous than quantitative questionnaires. It allows for the interview session to be more dynamic and adaptive to the potential changing circumstances (Bryman 2008: 465). Pilot testing enabled the researcher to make appropriate adjustments to the interview guide, which initially had been created out of extensive literature studies. The interview guide placed in the Appendix 1, formed an unconfined framework for all interviews with former child soldiers. Both open and closed

42

Martin Sjøen Master Thesis University of Agder ended questions were used in order to develop a greater understanding of the topic (Bryman 2008: 231-237). Probing and follow-up questions where used when issues needed more elaboration, or if answers given were too ambiguous. Interviews lasted between one and two hours, and notes were taken during the interviews. Efforts were made to minimize taking notes in an attempt to preserve the “natural setting” of the interview. Different interpreters where used through the course of investigation since most respondents spoke either Swahili or local languages. A total of twenty-four male former child soldiers were interviewed, whereas eighteen had passed through formalized intervention programmes, while six had no such affiliation (self-demobilized). Unstructured interviews were conducted with different practitioners and professionals affiliated with child soldier intervention programmes both in DRC and in Norway. These resembled informal discussions covering both established and new knowledge of importance to this work, and they proved valuable in designing the interview guide. In addition, several interviews were conducted by e-mail pre- and post the fieldwork. Participating interviewees in this research including former child soldiers and practitioners, derive from a myriad of different child soldier intervention programmes. Their approaches, perceptions and experiences may therefore differ accordingly. This research seeks not to scrutinize or evaluate individual programmes; rather it will attempt to highlight broader lessons learned, and challenges surrounding contextual child soldier reintegration.

3.3.2. Focus groups Focus groups are a widely used qualitative method in qualitative research, and it is a form of “group interview” where several people are conveniently interviewed together (Morgan 1997). This methodological approach can potentially take advantage of group dynamics and the communication that develops between participants. Ideally, the interaction in focus groups will challenge and stimulate participants to conceptualize and discuss ideas between each other, and it is useful method to explore people's knowledge and experience collectively (Bryman 2008: 473). Conversely, some limitations towards this approach include the sheer presence of a researcher, potential power-relations between participants, and external validity of these findings (Hart and Tyrer 2006). Focus groups were used in a supplementary way in this research, primarily to generate data on the perceptions of child soldier reintegration by local communities. A total of three focus groups were held in different communities in and around Bukavu. Participants where exclusively adults and each focus group involved five to eight

43

Martin Sjøen Master Thesis University of Agder persons of mixed gender. In hindsight, the combination of male and female participants may have implicated these focus groups, as male respondents had a tendency to dominate the discussions. Yet the decision to include mixed gender derived from advice made by a field contact, who argued that female homogeneous groups could arouse suspicion amongst male members of the communities. Duration of each focus group lasted between one and two hours. Topics of discussion evolved around the respondents’ perceptions of children in armed conflict, child soldier reintegration and “typical” behaviour of former child soldiers. The researcher who communicated through an interpreter had a facilitative role in terms of topics, and discussions developed in a relatively free manner.

3.3.3. Participant observation Participant observation is a method in which the researcher partakes in the daily activities, routines and events of a group of people, in order to learn the “explicit and tacit aspects of their lives, routines and their culture” (Dewalt and Dewalt 2002: 1). It is a frequently used method in qualitative research, and a common approach to collect data in the “natural” setting of the people being studied. In this research, participant observation was the least used method for collecting data, and initially not intended to be used at all. Child soldier reintegration is arguable such a complex and long lasting process, that participant observation would not be feasible within the restricted space of time. A proposition was however made by several contacts for the author to visit and observe various child soldier intervention programmes in activity. These intervention programmes included vocational training, cultivation and different social activities. One skill training program (tailoring and cultivation) for formalized and self- demobilized child soldiers was visited in Nyantende, South Kivu. Participants in these programmes included both males and females, and their ages ranged from early teenage to mid-twenties. Another program involved the distribution of meals to “street-children” in Bukavu, South Kivu. Recipients here were exclusively boys with reported ages from 8-17. This type of intervention work is typically referred to as a “non-targeted” program, where other war-affected children or orphans may receive assistance. A child soldier intervention program is a sort of “inauguration” of the socioeconomic reintegration. Potentially it may address some of the challenges faced by former child soldiers, yet it is not possible to assess the entire reintegration process based on brief visits to these programmes alone. The efficiency of participant observation is also linked with the number of observations a researcher does, i.e. the more observations the more efficient it

44

Martin Sjøen Master Thesis University of Agder becomes (King et al. 1994: 67). Thus the scientific value of this scarcely used method can surely be questioned. Another structural challenge of this method relates to the possibility of behavioral change of the participants, due to the presence of an “outside” researcher (Bryman 2008: 269). It does however offer the researcher some added insight into the thematic sphere of child soldier intervention.

3.4. Interpretation of data Data from the investigate process has been analyzed in the following way. Transcripts and field notes have been interpreted sentence by sentence in search for thematic key words. These key words are assigned to a series of open-ended codes. The codes are then grouped into concepts and classifications, eventually used to generate theories (Witzel 2000: 7). This analytical approach entails both inductive and deductive reasoning, and is commonly referred to as grounded theory (Bryman 2008: 541-550). Inductive reasoning is however presiding as this research is not constrained by established theories. Still, there are significant elements of deductive reasoning as the theoretical framework is based on existing theories. Grounded theory is frequently used in qualitative studies, and it aims to “neutralize the alleged contradiction between being directed by theory or being open-minded” (Witzel 2000: 1). Findings from this work may provide an insight to some of the many different nuances surrounding child soldier reintegration in the Kivus. Due to the contextual nature of research, a considerable challenge is related to proving the validity, or more specifically, proving the links between the different relations that are studied (Hart and Tyrer 2006: 32). Careful considerations must therefore be taken according to the external validity of this research, and it is important to be upfront with this at the initial stages of the work. This implication is not a new phenomenon in case study research, and it is necessary to consider whether it is possible to generalize the findings beyond the area of investigation at all (Blattman and Annan 2010: 896). It has been argued that the subject of child soldiering is so contextual that the “only generalization on the use of child soldiers is that there is no generalization” (Hart and Tyrer 2006: 33).

3.5. Limitations There are many limitations that could have prevented the entire, seriously hindered or simply challenged specific parts of this research. Besides the more common constraints such as insufficient time and lack of financial assets, there are other considerations when doing research in conflict zones. Security measures had to be taken both on a personal level, and for all others who were involved (Hart and Tyrer 2006: 14-15). A seemingly tense situation

45

Martin Sjøen Master Thesis University of Agder characterized parts of DRC which at the time of the fieldwork, recovered from a rather dubious presidential election. Road traveling between cities in the Kivus was largely discouraged, a situation which put serious limitations on the geographical accessibility. Cooperation with local organizations that provided first-hand knowledge and experience, was as mentioned required in this project. When working with other actors in research projects, it is vital for the scientific integrity that the research stays largely “independent”. As referred to in the section on sampling, respondents were largely selected by collaborating organizations, a structural weakness affecting the reliability indeed. Another sampling bias refers the isolated focus on former child soldiers. Ames argues that in order to understand the circumstances of child soldiering, researchers should conduct comparative studies between combatant and non-combatant children in armed conflict (2010: 17). Testimonies from current child soldiers are also absent in this research. Language barriers are another central challenge when conducting fieldwork (Borchgrevink 2003). This was experienced as transcripts from several interviews were scrapped due to inadequate interpretation and general uncertainty. Low levels of trust can hinder the possibility of conducting the research, and cultural bias may result in failure to capture or misinterpret the interviewee. Child soldiering is such a sensitive issue, and honest answers cannot always be expected as any “wrong” answer can potentially risk the respondents’ security (Andvig 2006: 10). Some respondents were also asked to reflect upon elements that occurred a long time ago, thus the accuracy of testimonies may be fragmented. Participants may on their hand provide answers which they think are beneficial for themselves or what they think the researcher “wants” to hear. Although development studies is an interdisciplinary subject comprising of many different sciences, this research suffered from a narrow focus involving basic socioeconomic factors. Research of this kind should ideally include several schools of thought including medicine, psychology, political science, history, economy and geography. The exclusion of particularly psychological and proper medical assessment is an undesirable implication.

3.6. Ethical considerations All research projects should according to Rye (2011: 4), consider ethical principles to minimize any potential risk for those involved. The share presence of a foreign researcher can potentially attract unwanted attention, arouse suspicion and change group dynamics within the community. Likely ethical issues must therefore be planned properly together with research partners. Ethical guidelines based on relevant literature (Brett and Specht 2004, UNESCO

46

Martin Sjøen Master Thesis University of Agder

2011, University of Minnesota 2003) was created to serve as the foundation for how the research was conducted, involving the utmost effort to ensure that no harm or stress was inflicted on interviewees13. Participation required general consent, and respondents were also informed on their ability to withdraw from the research up until a set deadline after their involvement (Bryman 2008: 121). Full confidentiality and anonymity are other ethical considerations that must be met (Hart and Tyrer 2006: 23). Due to the country's volatile situation, any correspondence or interviews with child soldiers, professionals, practitioners or local population residing or working in DRC, has been completely anonymized in this thesis. In accordance with an argument put forward by Barash and Webel, it is also important to be upfront with one's own values, which are among other liberal, antiviolence, pro human- rights and pro-social justice. Attempts to investigate conflict-related matters are most effective if build upon “open-minded and politically engaged intellectual efforts, including an attempt to understand all sides of complex debates” (Barash and Webel 2009: xiii). Furthermore, the role of the researcher is pivotal, and one should assess the appropriateness of different research methods for changing circumstances. It is important that the researcher avoids exercising any “therapeutic” roles. Respondents must be respected on the base of their individual personality and background, and participation must take place in their own “pace” (Hart and Tyrer 2006: 24). While it is generally recognized that voluntary expression of personal experience may help some respondents process sorrows and traumas, interviews must be designed not to do any harm (IASC 2007: 10). Lastly, it is important to reflect on what the “purpose” of this research is. Work of this kind should not be self-serving; rather it should have an aim of addressing specific challenges (King et al. 1994: 16). Practitioners and professionals should ideally be able to extract information to improve strategies and processes in their own operations. Participants might also have benefits by enhancing their abilities to reflect and express upon their own experiences. A clear notion of the purpose of this research should minimize the risk of it invading the privacy of those who have participated.

13 See Appendix 5. 47

Martin Sjøen Master Thesis University of Agder

Chapter 4: Findings and analysis 4.1. Disposition In the following chapter, the empirical findings from the fieldwork will be presented. It is divided into three parts. The first part provides a brief overview of the respondents’ general background. It is not based on any particular research question; rather it derives from the first research objective. Findings from this section includes rudimentary features of the respondents’ age, mode of recruitment, dispersal of armed groups, main duties performed, and how they exited from their respective armed groups. The second part is analytically based on the first research question, and it seeks to investigate specific socioeconomic traits of former child soldier when reintegrating back to society. Briefly included in this section is also two domains traditionally analyzed within the field of psychology. With that said, the domain of “hostility” avoids any systematic medical behaviors’ analysis, focusing instead on self-reported events and perceptions of hostility. Furthermore, the domain of “psychological well-being” is predominantly considered through existing literature from the current study area. This is augmented with a brief overview of self-reported psycho-social support the respondents have received. Overall, this chapter is based on the testimonies of the whole sample of twenty-four former child soldiers, organized around the four domains that constitute the conceptual framework exemplified by Annan et al. (2010). Additionally, the various interviews with field-experts and practitioners is included, as well as the focus groups with members of local communities. A summary of the findings in relation to the research question concludes this section respectively. The third part is based on research questions two and three. It builds partially on findings from the second analytical part. This is augmented with an analytical comparison between participation versus non-participation in child soldier intervention programmes. These elements will be analyzed according to the theoretical framework as described in section 2.7, in an attempt to assess the effects of child soldier intervention programmes in the Kivus. Further, a subsequent section is included to investigate the impacts of child soldier intervention programmes on the peacebuilding process in the Kivus. A summary concludes this section in light of the current research questions.

48

Martin Sjøen Master Thesis University of Agder

Part 1 4.2. Background information

4.2.1. Age dispersal At the time of interviews, all former child soldiers who participated in this research were male, each residing in or close to Bukavu. As mentioned, the exclusion of female respondents derived from an assessment on the ethical appropriateness. Many female child soldiers are believed to have experienced ghastly sexual- and human abuse mainly by men, and the author was therefore discouraged from including girls in the research. The average age of respondents was 23.1 years, with the youngest participant being 15 years, and the oldest 32 years. This makes median age 22.5 years, and mode age 22 years. Two out of the twenty-four participants were still underage during this research, with respective ages of 15 and 17 years each. Average age at the time of recruitment was 14.3 years. Median recruitment age was 15, while mode recruitment age was 17 years. The youngest recruited child soldier among the respondents had been 9 years, whilst the oldest had been 17 years at the time. Two respondents could not give precise account of their individual ages at the time of recruitment. Time served with the different armed groups ranged from 1 month to 8 years. Average time spent with armed groups among the respondents was 2.5 years, median time was 2 years and mode was 3 years. Three former child soldiers could not recite how long they had been affiliated with an armed group.

Categorizes Respondents age Recruitment age14 Time served15 Lowest 17 9 1 month Highest 32 17 8 years Average 23.1 14.3 2.5 years Median 22.5 15 2 years Mode 22 17 3 years Table 1: Age dispersal among respondents. Besides the large age-gap (seventeen years) separating the youngest and oldest respondent, it is useful to notice the overall corresponding ages of the interviewees. Average, median and mode age all place the respondents roughly within an age frame of 22-23 years, implying a somewhat homogeneous sample. On one hand, this may narrow the practical

14 Two non-response 15 Three non-response 49

Martin Sjøen Master Thesis University of Agder application of any theory generated from these findings, as by large the sample represents only a small segment of the former child soldiers in eastern DRC age-wise. Yet, the explicit focus may also strengthen the theoretical application of this age group located in “early” adulthood. Former child soldiers recruited between 12-17 years (adolescents) constitute 82% of the entire sample. Thereby correlating well with dominant theories regarding the majority of child soldiers being recruited in their adolescence (Hart 2008: 280, Psychology Beyond Borders 2008: 10). Yet the average recruitment age of 14.3 years presented in this work, is slightly higher than in comparable research. Comparable studies in DRC situate the average recruitment at approximately 12 years of age (Bayer et al. 2007: 557, ILO 2003: 37). Time spent with armed groups is dispersed from 1 month to 8 years among the respondents. Duration being a child soldier may generally vary greatly between individuals, as it is contextually related to each individual. In the study of war-affected youth in Uganda (SWAY), Blattman notes that abducted child soldiers' affiliation with armed groups ranged from “1 day to 10 years” (2009: 234). Two important considerations should be made in relations to the foregoing data on age and duration being a child soldier. Firstly, the sampling method and size of twenty-four respondents makes any findings tentative. They are therefore best viewed in an exploratory way. Secondly, perceptions of time including age and duration are largely Western concepts, and may not be appropriate in a Congolese setting (Drønen 2003: 7). The accuracy of these data are therefore debatable, yet its contribution towards this research is limited to providing minor background information.

4.2.2. The circumstances of recruitment Coercive means and physical abduction was the predominantly used recruitment method among the respondents. Only four out of the twenty-four participants cited they had voluntarily joined an armed group. This is considerably lower (17%) than comparable research, most notably the ILO study which found that two out of three Congolese child soldiers had some “initiative of enrolling themselves voluntarily” (2003: viii). At least three factors may elucidate the apparent gap between this study and previous work. First, perhaps the most obvious challenge relates to the possibility of sampling bias, and this factor is difficult to disregard. Second factor concerns the issue of expecting honest answers and the uncertainty of retrospective reflection by the respondents (Andvig 2006: 10, Honwana 2006: 14). Thirdly, it is also conceivable that the responses may have been decisively impacted by

50

Martin Sjøen Master Thesis University of Agder the majority of interviewees in this research being adults. Prior research has demonstrated that many former child soldiers acknowledge in retrospect how their preceding image of voluntary was somewhat deficient (Brett and Specht 2004, ILO 2003). Previously discussed in this paper is also how the notion of “voluntary” must be scrutinized. Perhaps speculative, but the cognitive development and abilities to critically reflect on one’s own background, should normally be higher among adults than adolescents and children. This may in turn account for why the majority in this sample view themselves as victims of forceful or abduction led recruitment.

Recruitment mode

17%

Coercive Voluntary

83%

Figure 8: Diagram on the mode of recruitment. During the fieldwork, the concept of voluntary recruitment was frequently discussed with various organizations, at which an interesting theory surfaced several times. A repeated claim was that there were indeed many child soldiers who had voluntarily joined armed groups in DRC. Yet, those who had joined voluntarily often continued their involvement with armed groups despite having options to refrain such affiliation. This aspect, although not confirmed in this research, would according to Brett and Specht indicate that the issues leading to recruitment had not been sustainably addressed (2004: 5). Despite the feasibility of such a hypothesis, it does not sufficiently explain the disproportionate low levels of voluntary recruitment found in this research, versus the higher levels in such studies as ILO (2003), and Bayer et al. (2007). The overall significance of this explanation is therefore questionable. As shown in the diagram below, recruiters of child soldiers from this sample involve a wide range of armed groups operating in eastern DRC. These armed groups include both

51

Martin Sjøen Master Thesis University of Agder regular armed forces (FARDC), Rwandese foreign militias (FDLR and Interahamwe), and irregular Congolese armed groups (RCD-ML, RCD, CNDP and Mayi-Mayi). The period of involvement for the former child soldiers is distributed throughout the entire period of contemporary armed conflict in DRC. Thus it involves participants from the First Congolese War (1996-1997), Second Congolese War (1998-2003), as well as the sub-conflicts that have occurred since 2003. One respondent did not provide information regarding the armed group he had previously been affiliated to.

Armed groups

Non-response

RCD-ML

FDLR

FARDC

RCD

CNDP

Mayi-Mayi

Interahamwe

0 1 2 3 4 5 6 7 8

Figure 9: Dispersal of armed groups.

4.2.3. Drivers of becoming a child soldier Findings from this research feature a wide range of drivers for becoming a child soldier. Based on the testimonies from the respondents concerning the prevailing use of coercive recruitment, the most central finding lends towards the recognized hypothesis that child soldier recruitment derives from a set of deliberate choices by armed groups (Andvig 2006: 17, Singer 2006: 135). These choices are undoubtedly connected with the periodical armed conflicts that have characterized DRC for nearly two decades, subsequently resulting in a great need for soldiers. Furthermore, these conflicts have most characteristics of a “protracted social conflict” (PSC) as defined by Edward Azar (1990 in: Ramsbotham et al. 2008: 84). They can also be seen as an extension on the notion of the “failed state” (Singer 2006) that is common characterization of DRC. By large, PSCs are featured predominantly in developing regions of the world. Another common trademark for these often relapsing conflicts is that they occur mainly within state borders, involving a high degree of civilian

52

Martin Sjøen Master Thesis University of Agder casualties (Ramsbotham et al. 2008: 86, Singer 2006: 38). All four respondents who were voluntarily recruited stated poverty as the main reason for joining an armed group. These respondents may therefore be linked with another leading hypothesis on how poverty is the most common denominator among child soldiers. In fact, all twenty-four respondents claimed they came from a financially strained background. Yet poverty in isolation does very little to explain why these respondents became child soldiers, when most other poor children in DRC do not. Among the voluntary recruited child soldiers in this sample, immediate families seemed less dismissive of their children becoming soldiers than those forcefully recruited. As will be highlighted in an upcoming section, some of coercive recruited child soldiers had their family or other relatives negotiate for their release. This may imply that social surrounding environment is a causative factor for child soldiers who voluntarily join armed groups. Besides these two drivers, parallels may also be drawn to other factors. Firstly, there are demographic factors such as the large access to children and adolescents in DRC (Britannica 2011). Secondly, there are cultural aspects that can be used to describe why especially adolescents partake in what is usually defined as adult activities. Lastly, there is the possibility that the high rates of vulnerable groups such as internally displaced may provide recruiters with relatively many prospective child soldiers (Achvarina and Reich 2010).

4.2.4. Child soldiers' roles One of the most disputable aspects of child soldiering concerns which roles and duties child soldiers usually performs (Wessells 2006: 8). Specific to this work, it proved difficult to distinguish what roles are most common, as the majority of respondents had in fact been assigned to a wide array of duties. In its definition of what a child soldier may be, the Cape Town Principles exemplifies roles as “cooks, porters, messengers and anyone accompanying such groups” (1997: 12). Yet, these roles are rather nonspecific, as few child soldiers, at least in this sample, had been exclusively a porter or a messenger. Hence the roles had to be classified into larger categories in order to make it more convenient and viable. Three different categories were therefore made, 1) soldier, 2) auxiliary support and 3) spy. A soldier was here defined as anyone who on a daily or regular basis carried a weapon and had experienced combat. Auxiliary support includes those conducting labour which may be described as “household tasks” (Andvig and Gates 2010: 2). Typically this refers to roles as cooks, porters and messengers. The role of a spy is perhaps self-evident, and in this research it involved mainly child soldiers affiliated with one armed group, who had then “voluntarily”

53

Martin Sjøen Master Thesis University of Agder join another armed group in an attempt to collect “strategic” information. The distribution illustrated in the diagram below, derives from the respondents own perception of what their main duties were. These roles are somewhat vague, and the boundaries between the categories are a bit unclear. For example, it might be difficult to distinguish between the roles of guards and soldiers. Many respondents have also performed different roles in their duration as a child soldier. Based on the empirical evidence, it is difficult to conclude in terms of the two leading discourses regarding the roles of child soldiers. As argued by Andvig and Gates, most respondents (46%) held rightly auxiliary roles as they would define themselves (2010: 2). Yet, this group is only modestly higher than those who considered themselves as soldiers (42.5%). If one includes the role of “spy” into the definition of “soldier”, as this task is undoubtedly related to a certain amount of risk, then this group would constitute the majority. Singer claims that most child soldiers are in fact soldiers, involved in various degrees of hostilities (2006: 76-77). Moreover, comparable studies regarding child soldiers in DRC reveal high levels of experienced hostilities (Bayer et al. 2007).

Child soldiers' roles

12,5%

41,5% Soldier Auxiliary support Spy 46%

Figure 10: Diagram on the roles of child soldiers. An overall image that is drawn from this section is that the internal structures within armed groups who utilize underage soldiers are dynamic. It seems not unlikely that newly recruited child soldiers are often set to do auxiliary duties in the early stages of their soldiering “career”. Then as time progresses, the child soldier may be trained for roles as guards, soldiers or spies depending on their cognitive capabilities, as well as the needs of the

54

Martin Sjøen Master Thesis University of Agder armed groups. The disputation regarding what roles the majority of child soldiers perform, i.e. “soldier versus auxiliary tasks”, is certainly interesting (Andvig 2006, Singer 2006, Wessells 2006). This is also important in regards to assessing child soldiers’ the combat experience. Still, by severely limiting the roles simply to whether they were “active” soldiers or not, this categorization may also undermine the dynamic nature of armed groups and their child soldier usage.

4.2.5. Departing the armed groups The majority (67%) of respondents in this research cited they had escaped from their respective armed groups. These escapees are evenly distributed throughout the various periods of violent conflict in DRC. This is moderately higher than the ILO conducted study, which found that “more than 50% of former child soldiers left the group on their own initiative” (2003: 48). Nonetheless, the high apparent number of escapees correlates somewhat with the high number of respondents (82%) who claim they were recruited by force or abduction. Perhaps self-evident, but one might expect the variables of recruitment and departure to be linked, at least to a certain degree. That is, child soldiers who are forcefully recruited may to a larger extent try and escape from their adverse environment. Interestingly though, two out of the four who claimed they had voluntarily joined an armed group, also stated they had escaped. The decision to escape could be what Blattman describes as an act of “self-determination and control”. People who in some cases witness or partake in “senseless acts of violence”, escaping can be seen as a crucial act of self-determination (Blattman 2009: 243). Still, with the sample being too small to generalize upon, these findings provide little more than speculative and exploratory data. All respondents left the armed group at least one year prior to the interview. Five of the twenty-four respondents said they had exited the armed groups through the army reunification process, or “brassage”, were the regular army and various armed groups was integrated into a new Congolese army (FARDC) (Baaz and Stern 2008: 63). While one of the five respondents could not recite precise account of when he was recruited and subsequently his departure, the reunification process transpired from 2002. Thus all of the five respondents had departed their armed groups by the end of the Second Congolese War, or in the period of the sub-conflicts from 2002 and onwards. A total of three respondents stated they had been set free as a result of third party negotiation with the armed groups for their release. Oddvar Espegren, Program Manager and Director of Christian Relief Network (CRN), explained how their work approach had altered over time. Initially, CRNs child

55

Martin Sjøen Master Thesis University of Agder soldier intervention program involved precarious negotiations with armed groups for the surrendering of underage soldiers. Later, armed groups who became a part of the army unification process (FARDC), would eventually transfer child soldiers increasingly on a voluntarily basis. Between 2003 and 2011, CRN as one of the most significant actors in DRC, received approximately 5400 child soldiers in their interim child reception centres.

Departing the armed groups

12,5%

Escaped 20,5% Army reunification Set free 67%

Figure 11: Diagram on the modes of departing armed groups. Eighteen of the twenty-four respondents (75%) had received reintegration support through intervention programmes. This is considerably higher than previous estimates, which state that only 50% of returning child soldiers “is believed to access reintegration support” (Coalition to Stop the Use of Child Soldiers 2011: 2). A rationale for the deviation in this work is that most respondents were directly appointed by staff members, thus one would expect a higher degree of respondents who have received reintegration support.

4.2.6. Re-recruitment Re-recruitment rates in northeastern DRC have generally been regarded as high. ILO for example, found that 60% of active child soldiers have belonged to “one or two” armed groups (2003: 37). Save the Children's study of demobilization and reintegration of child soldiers, discovered that as much as 69% of child soldiers had been re-recruited in parts of South Kivu (Verhey 2003: 54). A total of seven respondents (29%) in this sample claimed they had been re-recruited. All but one of these seven had been re-recruited by a different armed group, and they all stated that they had been forcefully re-recruited. In contrast with the

56

Martin Sjøen Master Thesis University of Agder

“initial” recruitment, were the superiority of Congolese child soldiers supposedly have some sort of initiative themselves (ILO 2003). Re-recruitment rates are believed to be almost exclusively through abduction and coercion. A study by Save the Children estimates that only 5% of re-recruited child soldiers are done so on a voluntary basis (Verhey 2003: 54). For those respondents who had been recruited more than once, the previous section regarding the circumstances of their departure relates to their latest affiliation with an armed group.

4.2.7. Summary of main findings part one Majority of the sampled respondents were adults with an average age of 23.1 years at the time of interview. Findings reveal that most respondents became child soldiers in their adolescence (14.3 years), complying with leading recruitment-theories (Hart 2008, Psychology Beyond Borders 2008). Time spent with armed groups varied greatly from one month to eight years. The large majority (83%) of respondents claimed they were forcefully recruited as child soldiers. This finding contradicts the few available data that from DRC (ILO 2003), yet this deviation may be explained by most respondents becoming more self-critical in their adulthood. Recruiters of child soldiers involve both regular and irregular armed groups from both Congo and Rwanda. Main drivers of child soldier recruitment can be understood as a set of deliberate choices made by the armed groups (Andvig 2006, Singer). Their incentives may furthermore be interpreted through theories of contemporary and protracted social conflict and lack of human development through the notion of “failed states” (Gates and Reich 2010). Yet for the four respondents who had voluntary joined armed groups, a mixture of contextual aspects such as poverty, culture and family are all causal factors. Most child soldiers hold a wide range of roles in their respective armed groups. Tentative findings suggest that as child soldiers advance through the structures of the armed group, they are trained for new and more comprehensive tasks. This may imply that the structures of armed groups are rather dynamic. Based on the evidence in this research and supported in relevant studies from DRC (Bayer et al. 2007), a significant part of the respondents perceive themselves as “soldiers”, reporting significant combat experience. The most common (67%) method of departing the role as child soldier was through escape, yet some were released through formalized army reunification process (21%), or set free as a result of third party negotiations (12%). Re-recruitment rates in this research were 29% which is considerably lower (69%) than comparable studies from DRC (Verhey 2003).

57

Martin Sjøen Master Thesis University of Agder

Part 2 4.3. Socioeconomic reintegration

4.3.1. Cultural barriers The scholarly ambiguity surrounding the concept of “successful” reintegration is evident, and it became apparent that the initial established term of the concept by the author was ill-defined, and conversely, rather problematic to use in this research. Tonheim writes extensively on challenges associated with how to define successful reintegration, including lack of participant perspectives, vague standards of measurement or the share ambiguity of the word “reintegration” itself (2010: 24). What turned out to be an obstacle in this research, involved the cultural barrier between preconceived Western ideas of socioeconomic reintegration, in contrast with a more “local” Congolese perception. Questions of how “well” child soldiers were integrated back with their families and local communities were often met with a sort of astonishment and sometimes even frustration by interviewees. Gradually it became clear that many viewed reintegration as an almost “physical” process. When the former child soldier had moved back to his immediate or extended family, he was reintegrated. The question of “how well” the former child soldier had been reintegrated seemed pointless. It was not until more detailed questions on different aspects including social acceptance, stigmatization, discrimination, livelihood and hostility were discussed as separate topics, that a larger impression of the reintegration process could be constructed.

4.3.2. Family reunification According to Wessells, one of the most painful aspects of war for most child soldiers is the separation from their families. Historic evidence suggests that child soldiers’ well-being is closely linked with the ability to reunite with their families (Wessells 2006: 183). Family reunion is for many child soldiers the prime enabler for putting war behind them, thus many practitioners and scholars view family reunification as the most important initiation to child soldier reintegration. Family reintegration in this section reflects only the perspectives of the former child soldiers in question, and it does not differentiate between those who participated in intervention programmes and those who did not. From this sample, the majority of former child soldiers had upon leaving the armed group, reintegrated back to their immediate families (parents), or extended families (distant relatives) if the parents were deceased. Only three out of the twenty-four respondents had been unable to do so. Of those three, two had been prevented because their families and relatives were entirely unwilling to accept them back and

58

Martin Sjøen Master Thesis University of Agder the third respondent because his family were believed deceased. These three respondents had therefore resided with acquaintances or on their own after exiting their respective armed groups. In the sample of twenty-four respondents, seventeen still lived with their family or relatives, while seven had established “own” families at the time of interview. Those who had established own homes were predominantly the older respondents. All respondents believed that family reunification was “necessary” for child soldiers when departing the armed groups. A forthright impression is undeniably a positive one with most respondents (88%) having been reunited with their families. This is likely one of the most crucial aspects in terms of a successful reintegration, albeit it only being the beginning of a long and difficult process (Wessells 2006: 183). There is little confirmable data on the percentage of former child soldiers in DRC who reunite with their families on a general level. Save the Children's review of 541 demobilized child soldiers in Bukavu, found that 77% had reintegrated with immediate or extended families (Verhey 2003: 59). Findings from the ILO conducted survey on the other hand, showed that only half the respondents in the research were able to return to the same families as they lived with before their recruitment (2003: 55). With that said, research from the African continent customarily show a high number of child soldiers returning home to their families. In the survey of war-affected youth in Uganda for example, Blattman describes a situation with the large majority (95%) being able to return home (2009: 237). Moreover, this has comparable features with case studies from Mozambique (Boothby 2006), and Sierra Leone (Humphreys and Weinstein 2004). A common denominator from these studies is that family reunification is often vital for the former child soldiers’ quality of life. This corresponds with the almost elated sensation many respondents described in this research, on being able to go home after their involvement with armed groups.

As explained by one respondent in this research:

I am very thankful that my parents accepted me back home. If they had not, I do not know where I would live.

4.3.3. Family acceptance Having established that family reunification is a central element in the reintegration process based on the participants perspectives; social acceptance is not always a present or constant factor in it. Accordingly, the relationship between the former child soldier and his relatives both pre and post recruitment, often determine the rate of the overall social

59

Martin Sjøen Master Thesis University of Agder acceptance (ILO 2003: 63). While the majority of former child soldiers in this research had to their great desire reunited with their closest families, at least in a physical sense, there were still prominent challenges affecting their social reintegration. Firstly, nineteen of the respondents (79%) reported having experienced periodical strained social acceptance, and varied levels of mistrust from members of their families or acquaintances. Usually this would materialize itself through what can best be described as periodically a “superficial” reunification. Despite having been reintegrated, the narratives from several former child soldiers depict a situation where they did not feel “completely” accepted by members of their families. With few points of reference, it is intricate to establish a universal understanding of what the word “completely” means, and this may vary between respondents. Yet levels of diminished social acceptance were frequently mentioned by this group of respondents. The “superficial” reunification seems more profound in the inceptive stages of the reintegration process, and 63% reported an increase in social acceptance over time. For example, older respondents who had been affiliated with armed groups around the period of the First Congolese War (mid-to late 1990s); spoke less of dismissive present-day relationship with their families than younger respondents. Moreover, many of the older respondents had evidently the “advantage” of not residing with their parents or other relatives anymore. Being more independent strengthened by a longer time to process and rebuild social structures, appears to help these older respondents reintegrate socially and avoid social conflict. Thus a tendency towards the hypothesis of social structures being improved over time can to a certain degree be found in this research (Annan et al. 2010, Blattman 2009). There were seemingly little difference between the respondents who had volunteered and those who had been abducted or forcefully recruited, as strained social relations were reported by the majority of former child soldiers in this work. Some authors imply that the mode of recruitment might be decisive for the level of social acceptance in the reintegration process (Blattman 2008b: 124). This means that former child soldiers, who voluntarily join armed groups, may find it more difficult to reintegrate socially. Although findings from this research challenge such theories, the limited sample of four respondents who voluntarily had become child soldiers, is imaginably too small to dispute this theory altogether. Another implication towards social reintegration was what some respondents described as a transformation in perceptions by their families. Based on the testimonies of the respondents, this was clearly ascribed to them having previously been child soldiers.

In the words of another respondent:

60

Martin Sjøen Master Thesis University of Agder

My parents look at me differently now. They do not talk to me in the same way as they did. They still care for me I think, but something has changed.

It appeared that the significance of this element differed greatly among respondents as ten interviewees (42%) described being treated differently or stigmatized on a more “regular” basis. Yet for those who described this as a “notable” problem, the respondents (seven in total) explained feelings of being “second-rate” family members. Some described having been unfairly treated, including experiencing relatively high levels of discrimination such as not being allowed to eat meals with the rest of the family. One respondent recounted how he had been denied going to school, while also having to do more labour than other members of his family. Despite such incidents, “only” two of the thirteen respondents voiced considerable malaise living with their families. Their housing alternatives and income possibilities were however severely limited, which in practicality meant that they had few options but to remain living with their families. In addition, the idea of family structures seemed to transcend the “individual”. Social relationship with family and relatives Yes No

Reunited with family or relatives directly after exiting armed group 21 3

Experienced some conflicting or incompatible social relations with family or rela- tives after reuniting with them 19 5

Feels regularly stigmatized and treated differently based on having been a child soldiers 10 14 Feels acutely discriminated and stigmatized based on having been a child soldiers 7 17 Feels that social acceptance has increased over time 15 9 Regards the present living condition with family or relatives as generally insup- portable 2 22 Table 2: Table on social relationship with family and relatives. From a liberal perspective, the fact that most respondents in this sample have in various degrees experienced strained social relations with their families and acquaintances is certainly worrisome. Still, how appropriate is this perspective in a Congolese setting? Only a relative small group (two respondents) described their social realm with families as adjacent to “unbearable”. This situation is certainly two-folded, and one can do little more than speculate towards these circumstances. For instance, do former child soldiers who experience strained social relations, tolerate them because the notion of “community” or group-belonging

61

Martin Sjøen Master Thesis University of Agder is arguably strong in regional parts of Africa (Bell 2007: 54). After all, the majority of respondents preferred reuniting with their families or relatives. Or is it the lack of satisfactory alternative living conditions as a result of widespread poverty, which subsequently prevents former child soldiers from establishing their own homes. The author holds the impression that the lack of satisfactory living alternatives constitutes the most significant reason, illustrated perhaps with those former child soldiers who had been able to establish own homes, describing less issues of social conflict. This is reinforced by the respondents living on their own describing being more independent then when they resided with families. Koch supports this notion as he claims that when former child soldiers are “married, have a home and have their first babies, they are settled and reintegrated in society” (Zimmermann 2010). When discussing family reunification, the relationship between the former child soldier and his family pre-recruitment should also be assessed. Wessells argue that many children become soldiers because of undesirable family situation in the first place, meaning that “family reunification is not universally applicable” (2006: 184). Regrettably, this research lacks comprehensive emphasis on the respondents’ family circumstances prior to their recruitment. Family relations that were strained before the children were recruited as soldiers are not likely to have improved by the time they reunite (ILO 2003: 55). It is therefore difficult to assess with certainty whether these strained relationships are exclusively the result of the respondents’ background as former child soldiers, or if past relationships also have an effect.

4.3.4. Active agency vs. victimhood – family reintegration Interestingly, when asked on what perceptions the respondents were left with regarding their families’ behaviour towards them, some actually described a sincere understanding of the varied social acceptance they had experienced. The change in social treatment many respondents had encountered was an understandable consequence of the violent conflict in DRC. The notion of belonging to a larger entity like family seemed so strong, that many respondents were willing to accept the actuality of these challenges. Thus it touches once more on the subject of individual well-being versus collective belonging, or “Ubuntu”, which may clash with liberal and Western perceptions (Kimmerle 2011). It may also give account for the cultural misunderstanding initially experienced in this work, when establishing the premises of successful social reintegration. While the share aspect of family reunification from the respondents’ perspectives may to the author be perceived as a “physical” or even superficial process. For the affected, it may in fact be the (re)-

62

Martin Sjøen Master Thesis University of Agder establishment of preordained social structures. Affiliation with a larger social sphere is viewed, at least contextually, as more appropriate both on an everyday level, but especially when trying to address social challenges. Thereby belonging to a larger entity like family may be regarded as an integral part of the reintegration/rehabilitation process. The respondents’ thoughts on how to approach the challenge of family reintegration yielded diverging answers. Many felt that family sensitisation would be both necessary and perhaps even sufficient to promote adequate social acceptance. Perhaps most important was the aspect of informing the affected families on the difficulties that most child soldiers had experienced. Wessells believes that these strategies can be beneficial (2006: 187). Several respondents felt that the circumstance around their recruitment and affiliation with armed groups had been an element beyond their control and aspiration. This can be linked with the high numbers who were recruited beyond their will in this sample. In point of fact, they considered themselves as victims equal to any other civilian in their community. Legally, former child soldiers are victims by definition regardless if they volunteered or not16. A significant challenge seems therefore to be linked with conveying this actuality to their families. Some respondents believed that they too had some responsibility in becoming socially accepted by their families. Any former child soldier who wanted to return to his family should also earn their respect and prove he has well-meant intentions. This perspective, expressed by a smaller part of the sample (21%), illustrates a profound sense of responsibility. It is seemingly placed far from the more common perspective of “victimhood”, yet these two perspectives are not mutually exclusive. In all probability, elements from both these approaches contain important features for achieving a sustainable family reintegration. Ultimately, successful reintegration should include transformational efforts by all actors to (re)-establish social structures.

4.3.5. Community reintegration - child soldiers’ perspective Discussed in a previous section is how prolonged armed conflict is impacting most aspects of society. Contemporary civil war has in terms of casualties and damages increasingly been shifted from targeting primarily soldiers to civilians. One of many authors, Dallaire notes the brutal irony of the term “civil war”, as there is surely nothing civil about it (2011). Not only do civilians carry the heaviest burden during armed conflict, they also have

16 The issue of criminal responsibility for children in armed conflict is a contested one (McMahan 2010). According to the Convention on the Rights of the Child (CRC), the best interest of former child soldiers must always be the primary consideration (Grossman 2007: 8). As of 2009, 194 countries have ratified the CRC. 63

Martin Sjøen Master Thesis University of Agder to contribute the most in reorienting societies from war to peace in the aftermaths. This includes the key issue of disarming and reintegrating former child soldiers. According to Sendabo, communities affected by war know very well the needs for rehabilitation and social reintegration, yet it is not clear if “they know how to deal with the problem itself” (2004: 64). It is debatable whether lack of effective measures to conduct peacebuilding is a result of little “knowledge”, as grievance, hatred and mistrust can all implicate the restoration of peace. The ability to exonerate and accept if not forgive a former “enemy”, is undoubtedly a momentous yet important task in restoring and sustaining peace in a country that has barely seen any in modern history. Community reintegration in this section reflects solely the perspectives of former child soldiers, while the perspectives of local communities are included in a succeeding section. A total of sixteen respondents in this research had moved back to their local communities after exiting the armed groups, while eight respondents had relocated to other villages. The latter option includes the three former child soldiers who had been unable to reunite with immediate or extended family. Findings from this research suggest that community reintegration is likely a more strenuous task than family reintegration. Yet it should be noted that the term “community” is nonspecific and perhaps too generalized. It implies a homogenous group who exhibits a similar or identical behaviour, which simplifies the reality.

4.3.6. Community acceptance All twenty-four respondents had experienced considerable challenges when being reintegrated into their local or host communities. Apparently, it played little role whether they had relocated to other communities with their families, as the stereotype of being a “kadagos” (child soldier) was often persistent. Challenges faced by former child soldiers include frequent events of discrimination. Examples were given of having opportunities of livelihood hampered by a preconceived idea of them being less noble and trustworthy. The majority of respondents had also experienced abusive name calling and malicious gossip aimed at them. They were often accused of having committed criminal acts, especially unsolved incidents. According to the former child soldiers, this treatment was unsubstantiated as the respondents claimed they behaved no differently than other members of society. These findings mirror somewhat other studies from DRC. The ILO conducted survey in central Africa found also high levels of prejudice from local communities towards former child soldiers (2003: ix). In her research, Tonheim investigated levels of stigmatization among former child soldiers in eastern DRC. Although her focus was exclusively on females, her respondents reported

64

Martin Sjøen Master Thesis University of Agder frequent occurrence of social rejection and stigmatization (Tonheim 2011). Several studies from other African countries have on the other hand reported findings that are modestly contradictory to these. Research in Mozambique (Boothby 2006), Sierra Leone (Humphreys and Weinstein) and Uganda (Blattman 2009), suggests that former child soldiers usually find social acceptance in their communities. Comparable to the process of family reunification, community reintegration seemed to improve slightly over time. Fourteen respondents (58%) felt that the situation had become better. Evidently there was also correlation between time since reintegration and decreasing levels of social “difficulties”. Despite this apparent improvement, twenty respondents (80%) had in actual fact experienced recent acts of prejudice, stigmatization or discrimination in their local community. “Recent” was here defined within the last few weeks. None of the respondents had participated in what is often referred to as traditional “cleansing ceremonies”, were child soldiers perform symbolic rituals to cleanse their spirit. Comparable studies have found such rites to be favorable for male former child soldiers when reintegrating back to society (Blattman and Annan 2008a: 19, Boothby 2006: 244, Sendabo 2004: 77). Traditional rites are generally believed to be used among a significant proportion of rural Congolese (Boya 2010). Even the use of child soldiers in DRC is surrounded by myths and beliefs of magic. Especially the Mayi-Mayi associate children with “kindoki” (witchcraft) such as being immortal in combat (Coalition to Stop the Use of Child Soldiers 2010: 4). Yet there is little knowledge on the use of traditional rites when reintegrating former child soldiers in the context of DRC. According to one practitioner, traditional rites do occur, yet the importance and scope varies between cultures and tribes. Lastly, the vagueness of the word “community” in this setting should be assessed. Most respondents did not describe a situation where the community as an entity collectively discredited them. Rather they gave the impression that those who displaced discriminative, prejudice or stigmatizing behaviour, was limited to a handful yet significant group of “perpetrators” within the village. Yet it is apparently significant enough to affect the social well-being of the respondents, and consequently, should be taken earnestly.

4.3.7. Addressing communal inequity While the common belief among most respondents was that “degraded” social treatment from members of local communities was unwarranted, living with family or relatives provided the former child soldiers with additional social support. Belonging to a larger social entity like family helped remove social inequity, and show that former child

65

Martin Sjøen Master Thesis University of Agder soldiers could function like any other member of the community. Gaining acceptance from members of the local community was described as integral by most respondents. Thus the process of family- and community reintegration seems interlinked, and the level of success of either may be somewhat dependent on the other. Furthermore, the process of family- and community reintegration shares many similarities such as relative high prevalence of social discrimination, prejudice and stigmatization. At the same time, these social implications and challenges seems to diminish significantly for most respondents over time, as shown in comparable studies (Annan et al. 2010: 6).

Community reintegration Yes No

Reintegrated into local community after exiting armed group 16 8 Experienced prejudice, discrimination or stigmatization from members of local communities 24 0 Experienced degrading social treatment recently 20 4 Experienced increased community acceptance over time 14 10 Table 3: Table on community reintegration. A difference between family- and community reintegration in this research however, was that some respondents showed an “understanding” for the change in perception or attitude towards them from their respective family members, but not from the local community in question. Altered communal behaviour towards former child soldiers was nothing less than unjust according to the respondents, and little emphasis was put on the former child soldiers’ ability or intended wish to reintegrate themselves. The notion of victimhood seemed strong as the common claim was largely that “successful” community reintegration was dependent exclusively on the community’s behaviour, rather than the former child soldiers themselves. Regarding measures to address these social injustices, the notion of community sensitization was the predominant “proactive” tool mentioned by the respondents. Yet several of respondents believed also that the situation would naturally improve over time. Most respondents sincerely believed they would eventually be well integrated in their communities.

4.3.8. Community perspective on child soldier reintegration Community attitude(s) towards the return of former child soldiers is probably of equal importance to that of family acceptance (where it is possible), at least in a long-term perspective. Social interaction and economical livelihood are dependent on an inclusive community. The three focus groups organized around this topic generated quite diverging

66

Martin Sjøen Master Thesis University of Agder perspectives to the extent that it is difficult to generalize upon its findings. However, a common impression among many community members was the individual families had to “decide” whether a former child soldier could return to his village. If the families were unwilling to do so, then the former soldier should not go “home” to his community. This perspective is comparable to that explained by former child soldiers, who also characterize family as the decisive factor in the reintegration process. It mattered only modestly which armed group the child had been affiliated to, with most respondents displaying a general resentment towards all armed soldiers in DRC. Still, child soldiers belonging to armed groups originating or receiving support from foreign countries (Rwanda), came somewhat across as less welcoming than others. This may derive from a fairly common perception that foreign actors are responsible for much of the torment and civil abuse many Congolese experience (Boya 2010). There was a modest divide whether the mode of recruitment would affect community acceptance. As stated by a community member:

If someone is abducted by soldiers, he can still choose to run away from them.

Most however, argued that the mode of recruitment was insignificant, and accepting children to come back was largely a community responsibility. Nearly all respondents felt that the circumstance of having children and adolescents participating in violent conflict was exclusively an adult-made decision. In other words, they recognized that the situation of armed conflict and subsequently, the deployment of underage soldiers were beyond the child soldiers’ control. Therefore it should preferably not be held against them. More importantly however seemed the actual background of the soldier himself. This subject which concerns the moral or legal responsibility of child soldier is a challenging aspect, dividing scholars apart. A very common view emerging from the focus groups was the child soldiers who had committed severe atrocities, should not be acquitted on the circumstance of them being underage. While they apparently acknowledged a certain amount of victimhood, they also argued that child soldiers should be held responsible for what they had done. Accordingly, this accountability included criminal acts committed outside armed groups, as many respondents believe former child soldiers often perpetrated such acts.

A question many asked:

67

Martin Sjøen Master Thesis University of Agder

Why should we forgive someone who has hurt our people and continue to do so?

Personal security was also mentioned as an implication towards community reintegration. Some respondents felt that most soldiers constitute a significant risk towards the fragile peace, and there was genuine worry regarding their “aggressive nature”. In its own respect, this is undoubtedly a compelling argument as many civilians have experienced severe abuse by soldiers in recent years. To recap, it is difficult to summarize the prominent perspectives from participating members of local communities as they deviate substantially. Moreover, this topic is given a more comprehensive focus in the succeeding section. A broad summary however, the determining factor for community reintegration of former child soldiers seems to depend chiefly on his family willingness to accept him. Furthermore, whether or not the individual child soldier has committed severe atrocities seems also to matter.

4.4. Conceptualizing hostility According to Blattman and Annan, former child soldiers are commonly depict as “traumatized, socially excluded, and violent” (2007: 2). While such stereotypes are frequently used, the empirical evidence of these matters is rather inconclusive. At the same time, studies suggest that local populations often fear former child soldiers returning to their communities (ILO 2003: ix, Wessells 2006: 182). It is therefore likely that hostility is an important issue in the process of child soldier reintegration. Hostility is here defined by Merriam-Webster as either a “deep-seated usually mutual ill will”, or “hostile action” (2012a). Some theories suggests that factors like 1) past experience of interpersonal violence, and 2) a customization of accomplishing objectives through force, may increase the likelihood of committing violent acts in a post-conflict setting (Annan et al. 2010: 6).While the potential link between “traumatic” experience and violent behaviour is interesting, it might also be a speculative subject lacking affirmative knowledge. Evidence seems also to suggest that hostile behaviour decreases over time (Boothby 2006: 249). In the context of DRC, scarcely any research on this topic is at hand. Among few is the clinical-psychological non-randomized study of a unit former Congolese and Ugandan child soldiers. One of its main findings is that respondents who show relevant symptoms of post-traumatic stress disorder (35%), also reveal less “openness to reconciliation” with “significantly more feelings of revenge” (Bayer et al. 2007: 558). Feelings of revenge can be characterized as hostile behaviour, and the effect of psychological trauma should be considered when former child soldiers are reintegrated into 68

Martin Sjøen Master Thesis University of Agder civilian society (Bayer et al. 2007: 555). PTSD may subsequently be an important factor in explaining post-conflict violence, yet it should not be mistaken for violent acts itself. In fact, the linkage between these factors is not sufficiently proven. Moreover, former child soldiers who suffer from PTSD or any other war-affected malaise do usually not constitute the majority of this group altogether (IASC 2007: 11). Hence, there is a need for clinical- psychological evaluations to be combined with reliable data concerning acts of hostility and violence. With the main focus of this study being social aspects rather than psychological ones, the predominant focus was to investigate personal perceptions and experienced acts of violence from former child soldiers and local communities. In other words, in-debt clinical- psychological analysis of human behaviour is absent in this section. Another central concern is that answers are solely self-reported and retrospective, which in turn can affect the perspectives.

4.4.1. Hostile behaviour among former child soldiers Former child soldiers interviewed in this research showed few signs of hostile attitude or general animosity towards other people in general. In fact, the majority renounced acts of violence such as physical fights, disobedience or verbal threats altogether. Perhaps surprising considering the levels of adult-led mistreatment several of them may have experienced. However, some respondents had been involved in what they perceived was unprovoked hostility towards them, which will be further explored in an upcoming section. Most respondents gave an impression that they would like to put their soldiering life behind them, and everyone claimed that they had not used violence as a means to an end since being reintegrating. Comparable studies from Uganda (Blattman 2009: 236), Mozambique (Boothby 2006: 245), and Sierra Leone (Humphreys and Weinstein 2004: 4), have similar findings. There were some incidents of reported resentment towards Congolese politicians, foreign intervening armed groups (mainly Rwandan), and the national Congolese army (FARDC), for their “hostile and corrupt” demeanor. Opinions like these are however not rare amongst most Congolese civilians, and it is doubtful whether they are linked to their individual background as former child soldiers. Speaking to the newspaper the Deutsche Welle about child soldiers, the director of an intervention program in DRC, Achim Koch says:

They’re very peaceful... very peaceful. All of them have been forced to go to the army – to the miliz. That means they are very lucky that they came out of it. And now they get a possibility to find another existence in society besides being a soldier (Zimmermann 2010).

69

Martin Sjøen Master Thesis University of Agder

Similar statements appeared in interviews with practitioners and field-experts in DRC. Namely that child soldiers who had been reintegrated back with their families and local communities did not show higher levels of hostility than other war-affected people. There was however a claim that the initial reintegration process could occasionally be challenging. This period often implies the “onset” of the reintegration process where child soldiers are either forcefully or voluntarily transferred from armed groups to a civilian setting (demobilization), sometimes with a relative lengthy stay in a transit centre. Aggressive or unwarranted behaviour is supposedly not uncommon by some former child soldiers in these environments. Oddvar Espegren gave an example of how their work in Northern Kivu had experienced hostile acts from former child soldiers. Their transit centre in Beni (mainly made of straw huts) had on several occasions been burned down by former child soldiers dissatisfied with the intervention program. Not been given “enough meat” was cited as a triggering factor for the disgruntled behaviour at least on one occasion. Still, Espegren claims that few former child soldiers quit the intervention program prematurely or escaped from the transit centre overall, and that most participants adapted into the system 17 . Perhaps this implies that challenges of aggressive behaviour are of a transitional nature limited to a few former child soldiers, and that they can be dealt with through preparatory work. Wessells argue that there will always be a “few bad eggs” among former child soldiers, yet experience show that sufficient recovery assistance and family reunification will often reduce hostile behaviour (2006). With demobilization being the very start, there is little evidence that hostile behaviour occurring in this initial phase will maintain throughout the entire reintegration process.

4.4.2. How local communities perceive former child soldiers Contrary to the claim from all former child soldiers in this study denouncing hostile behaviour, focus groups held in local communities showed diverging perceptions by different stakeholders. A common impression among some of these participants was that many former child soldiers displayed acts of both “violence and crime”. None of the questioned had personally experienced aggressive incidents by former child soldiers, yet it was a prevalent perception that many soldiers were unethical people. To the author’s understanding, this hostile perception was not limited to child soldiers in isolation, but to all soldiers both regular/irregular and national/foreign. The extensive nature of this view was largely confirmed through an interview with a university lecturer in Goma. He claimed that the relationship between civilians and Congolese authorities like police and army had been

17 Espegren, O. Program Manager and Director at Christian Relief Network (CRN): Lecture in Stavanger 14.03.2012. 70

Martin Sjøen Master Thesis University of Agder strained since the period of Mobutu. Armed authorities had always been deliberately paid low wages by the government, if they were even paid at all, evidently resulting in widespread corruption, extortion and misuse of their position of power to earn money. In their attempt to make sense of the violence in DRC, Baaz and Stern found similar views regarding the relationship between soldiers and civilians (2008). Both former and active soldiers often become scapegoats for the many problems in DRC. Is the perception by local communities on the moral deficit among child soldiers entirely a constructed one with little base in reality? Anecdotal evidence might suggest so, yet it is impossible to blatantly refuse the actuality of such a perceptions altogether. Former child soldiers interviewed in this research were all reintegrated into a community, most residing with immediate or extended family in what could be described as “average” living conditions. Thus this sample excludes those unable to socially reintegrate, potentially becoming homeless or residing in other poor living arrangements. Participant observation of “street-children” in Bukavu through a basic support program provided a brief insight towards a group of less fortunate war-affected children. This group was not exclusively former child soldiers, yet their strained everyday situation revealed the potential harsh consequences of not having adequate social support available. The staff associated with this program claimed that the majority of street-children were indeed a vulnerable group, surviving largely on begging, alms and petty theft. Dependency on drugs and alcohol was apparently not uncommon either. Yet these features are not limited to vulnerable children in DRC as it can probably be found among street-children around the world. Nor do former child soldiers exclusively make up this vulnerable group. While the dissipated perceptions towards former child soldiers found among several respondents is perhaps not completely unwarranted, it is likely more the result of misconception based on the behaviour of a minority of former child soldiers.

4.4.3. Hostility towards former child soldiers Despite the seemingly conflictual views regarding hostility exhibited by former child soldiers, several of the former child soldiers actually reported being the recipient of hostile acts. A total of seven of the twenty-four respondents claimed they had experienced unprovoked violent attacks on the basis of them being former soldiers. Perpetrators of such acts included both members of their local communities and local authorities such as soldiers and police. It appears that hostility towards former child soldiers can be an equal if not even a larger implication towards the reintegration process, than hostile acts committed by former

71

Martin Sjøen Master Thesis University of Agder child soldiers. Yet this feature is hardly considered in relevant literature.

One respondent described the following incident:

I was with another former child soldier and we were travelling to his village. We spent one night in a different village, and the people there knew who we were. During the night we heard much noise outside and there were many people shouting at us to come out from where we were sleeping. We were scared and tried to escape but could not get away. They beat us up pretty badly.

Albeit being rather exploratory, the findings of experienced prejudice, stigmatization and hostility described by many of the respondents are surely alarming. It also questions whether the prevailing practical and academic focus on post-conflict hostility is adequately addressed in the first place. The almost exclusive focus on former child soldiers’ hostility, excluding any behaviour of local communities, provides only one dimension of this domain.

4.5. Economic livelihood The success or failure of child soldier reintegration is often presumptively measured on the base of the participants’ ability to obtain sufficient means of income, thus truly becoming “independent”. This is tied to the devastating effect armed conflict has on economic and human development (Collier 2008). Within the field of child soldiering, economic livelihood is increasingly recognized as a crucial element for the former combatants’ well- being (Annan et al. 2010: 3). This is apparent through the study of both human and physical capital. Thus the domain of “economic livelihood” should presumably be conceptualized in a wider sense than merely encompassing income generating work. In this research, it comprises work, education, or other “meaningful” activities as expressed by the participants. There is little apparent doubt concerning the importance of this domain based on the perceptions of the respondents themselves. In fact, all former child soldiers interviewed in this research stated that economic livelihood was the single most important aspect regarding their own reintegration process. Challenges related to social acceptance, discrimination and hostility, was deemed as less significant than that of accumulating income. This perception applied to all respondents, and there was seemingly no difference in terms of age or social situation. A notable observation was that only three respondents characterized themselves as being able to sufficiently support themselves financially.

72

Martin Sjøen Master Thesis University of Agder

Occupational distribution

8% 8% 4% Business owner 12% Carpenter Cultivator 16% Brick layer 42% School 4% Unemployed 4% Vocational training Social worker

Figure 12: Diagram on the dispersal of occupations. Of the twenty-four respondents, twelve (50%) were engaged in different jobs. Ten respondents (42%) were without regular income-generating work (here defined as unemployed), depending largely on support from families and relatives. Two respondents, among the youngest in the sample, attended school and vocational training respectively. Similar to those characterized as unemployed, respondents receiving education were dependent on the support from family and relatives. Poverty is by no means isolated to include ex-combatants in DRC, a country ranked as one of the most fragile and poorest in the world (African Development Bank Group 2008: i). A unique feature in DRC is argued to be the large informal sector, which during Mobutu’s rule accounted for more than three times that of formal sector (Koyame and Clark 2002: 207). Conversely, much of this tradition has been impaired in the course of two decades with armed conflict. Besides poverty, unemployment rates are also soaring. Yet based on these findings, former child soldiers interviewed in this research (42%) seems relatively worse off than the general population (29%) in the comparable region (Eastern Congo Initiative 2011: 304). The empirical evidence in this research corresponds largely with other interdisciplinary surveys conducted on the African continent. Case studies from Liberia (Sendabo 2004), Mozambique (Boothby 2006), Sierra Leone (Humphreys and Weinstein 2004), and Uganda (Blattman and

73

Martin Sjøen Master Thesis University of Agder

Annan 2007), all reveal similar findings regarding the importance and struggles of economic livelihood. Moreover, as a theme it has also gathered increasing recognition from the medical/psychological paradigms (Mels et al. 2010, Psychology Beyond Borders 2008).

4.5.1. Economic impact of armed conflict on child soldiers A common view among most respondents is that their background as former child soldiers is a tenable explanation for their apparent economic challenges. The social mark as a former soldier implicates their job opportunities. Even the three respondents who characterized themselves as economic viable in present context, acknowledged the additional burden of having been associated with an armed group in the past. This finds relevance with certain theories, which suggest that a soldiering background may be associated with lower “human capital, wages and wealth” (Annan et al. 2010: 3). Thus the factors of social and economic reintegration are closely interlinked. What is more, most respondents believed that if given a chance, having a meaningful occupation would help reduce potential social challenges, ultimately demonstrating the “good intentions” they claimed to represent. Without trivializing this perspective, a fundamental question remains whether former child soldiers are statistically worse off than their non-combatant peers. The reasoning presented here is solely based on self-reported perceptions from the former child soldiers. In other words, there is an apparent void of comparative data between former child soldiers and other war-affected youth. One of the largest research gaps in the field of child soldiering relates to this very phenomenon (Ames 2010: 22). With nearly 90% of the population in eastern DRC surviving on less than $2 a day (Vinck et al. 2008: 22), poverty is certainly endemic and not isolated to former child soldiers. The role of economic livelihood within the philosophy of conflict transformation will be further assessed in part 3 of the analytical section.

4.6. Psychological well-being The following section examines current studies on mental well-being among war- affected youth in DRC. It is almost exclusively a literature study, and the rationale to include a psychological domain in this analytical section, is to provide a more comprehensive picture on the contextual circumstances of child soldier reintegration. According to IASC, mental disorders make up 40% of the leading causes of disability worldwide, yet “mental health is among the most under-resourced areas of health care” (2007: 123). Psychological theories regarding child soldiers have very briefly been discussed in this thesis. To the authors’ knowledge, there is only one peer reviewed article dealing exclusively on the mental well-

74

Martin Sjøen Master Thesis University of Agder being of child soldiers containing empirical evidence from DRC. That is the clinical- psychological study on child soldiers in DRC and Uganda by Bayer et al. (2007). Additional research is therefore supplemented by other studies regarding traumatic war-related exposure among adolescents in northeastern DRC. Emerging studies have increasingly enabled experts to assess the situation for war-affected youth in this region. Most of these studies however, are exploratory of nature, characterized by “small sample sizes, unrepresentative samples, or an absence of control groups or causal identification” (Annan et al. 2010: 4). From a Western perspective, some of the trends found in these studies are unquestionable disconcerting.

4.6.1. Exposure to distress factors Traumatic exposure such as witnessing or partaking in violent acts and other human- rights abuses are commonly acknowledged as a frequent cause of trauma or PTSD in conflict areas (Psychology Beyond Borders 2008). The numbers of children and adolescents who have witnessed or experienced violence in DRC are arguably very high, undoubtedly leaving psychological marks on many. A study in the northeastern Ituri district bordering North Kivu, revealed that 95% of the sampled adolescents between the ages of 13 and 21, had experienced at least one (nearly five on average), potential traumatic event. The most frequent traumatic exposure (72.4%) being the murder of a family member or friend (Mels et al. 2009: 528-529). Similar findings were discovered among internally displaced adolescents in Ituri district (Mels et al. 2010). Nearly one-quarter of the sampled adolescents in northeastern DRC had experienced being forcefully separated from their family at one point. Furthermore, vulnerability factors such as poverty and Figure 13: Map of northeastern DRC. displacement are cumulative, and Source: BBC (2009). believed to increase the association of traumatic exposure (Mels et al. 2009: 528, 2010: 1101). Sexual exploitation of adolescents is reported among 9.7% of a sampled unit in Ituri district (Mels et al. 2009: 528), or in every 6.8% household among respondents in eastern DRC (Johnson et al. 2010: 559). There are no available figures on substance abuse applicable

75

Martin Sjøen Master Thesis University of Agder specifically for children or adolescents, yet 46.3% of men and 21.8% of women were reported to be current substance abusers in a study conducted in eastern DRC (Johnson et al. 2010: 559). In the group of Congolese child soldiers, traumatic exposure may be even more widespread than among their non-combatant peers. More than 90% of a sampled unit of child soldiers in Goma and Bukavu claimed they had witnessed severe violence and/or murder. Acts of sexual abuse applied to almost one-third of the interviewees, while nearly 80% had been seriously beaten. Furthermore, among the 111 Congolese child soldiers participating in this survey, 64% claimed they had personally killed someone (Bayer et al. 2007: 558). Lack of comparable medical/psychological studies hampers any attempts to check for consistency in these findings. Yet the ILO conducted study discovered similarly high levels of potential traumatic exposure (2003). Figures from the studies conducted in DRC (Bayer et al. 2007, ILO 2003) are perhaps moderately higher than comparable studies in other African countries. Similar tendencies to these mentioned are nonetheless found in Liberia (Sendabo 2004: 43), Mozambique (Boothby 2006: 248), and Uganda (Blattman 2009: 234).

4.6.2. Symptoms of psychological distress All studies that make up the previous section reveal high prevalence of psychological distress, likely as a result of the equally high levels of traumatic war-related experiences. Approximately 35% of the child soldiers showed relevant symptoms of PTSD in a medical study Trauma: Psychic or behavioral condition (Bayer et al. 2007: 555). This is actually usually following severe considerably lower than the study of war- mental/emotional stress or physical injury. affected youth in Ituri district, which found that Source: Merriam-Webster (2012b). 52.2% showed symptoms of PTSD (Mels et al.

2009: 525). Another study in eastern DRC, PTSD: although sampling war-affected adults, found Emotional condition usually following a traumatic event. It sometimes that 50.1% met symptom criteria for PTSD involves recurrence of trauma, (Johnson et al. 2010: 553). Based on leading feelings of intense fear, sleep deprivation, anxiety or other theories, it is perhaps contradicting that the study distresses. with the highest levels of traumatic exposure Source: Britannica (2012). show the lowest level of PTSD symptoms (Annan et al. 2010: 4). However, this may be related to sampling bias or other methodological challenges. In addition, the psychological

76

Martin Sjøen Master Thesis University of Agder effects of child soldiering various greatly between studies and countries (Psychology Beyond Borders 2008: 14-16). As discussed, symptoms of PTSD among child soldiers may be associated with increased hostility such as less openness to reconciliate and more feelings of revenge (Bayer et al. 2007: 555). Respondents showing clear symptoms of PTSD may also experience serious trauma, depression, anxiety and acute nightmares. In Western medicine at least, it is customarily regarded that people suffering severely from the mentioned psychological distresses will often require specialized and individualized mental health care (Psychology Beyond Borders 2008: 37). Nevertheless, the appropriateness of Western approaches in non-Western settings is a current and contested topic in this field. Besides stressing the need for contextual appropriate psychological support, empirical evidence suggest also that restoration of socioeconomic factors can be considered as an important starting points for successful healing (Mels et al. 2010: 1102).

4.6.3. Received psycho-social support Among the eighteen former child soldiers who participated in intervention programmes in this study, none had been offered specialized/individual psychological evaluation or counseling. Nine participants (38%) had received other types of psycho-social support, mainly organized group conversations with former child soldiers and social workers.

Received specialized support

Received general medical assistance Yes No

Received basic psycho-social support

Received basic training

0 20 40 60 80 100

Figure 14: Diagram on psycho-social support. 77

Martin Sjøen Master Thesis University of Agder

It did not seem imperative for most participants in this research to receive more specialized psycho-social support, at least not in the initial reintegration phase. Some voiced a notable complaint regarding the complete lack of long term follow-up support. Only three respondents (13%) said they had acquired general medical (somatic) assistance from a doctor in the form of physical examination. Yet nearly all the respondents had received basic guidance in children’s hygiene, sanitation training and sexual safety strategies. None of the six respondents who self-demobilized had received any of the mentioned assistance, and had therefore relied solely on family and community support. To a large extent, practitioners interviewed in this research confirmed that a substantial part of Congolese child soldiers are exposed to potential traumatic events. Further, this finds support in the relatively high number of child soldiers who claims they have been participating in active fighting found both in this work and comparable studies (Bayer et al. 2007, ILO 2003). Type of trauma Girls Boys Total

1: Has witnessed violent acts against family members or friends 19.7% 23.8% 22%

2: Has seen someone being raped 35% 31.8% 33.2%

3: Had family members or friends violently killed during the war 70.4% 74.1% 72.4%

4: Has seen someone being killed 59.5% 70.9% 65.7%

5: Has been injured during the war 12.3% 19.6% 16.3%

6: Has been sexually abused 10.5% 9% 9.7%

7: Has been kidnapped by an armed group 12.6% 23.6% 18.6%

8: Has been forced to kill, injure, or rape someone 3% 9.8% 6.7%

Table 4: Table on traumatic exposure. The table above replicates some of the findings from a cross-sectional survey conducted on the traumatic exposure in northeastern DRC (Mels et al. 2009). Respondents in this study were adolescents and young adults aged 13 to 21. These findings reveal that traumatic exposure is widespread in this region, potentially increasing the risk of developing post-traumatic stress symptoms. They may therefore be indicative of an immense need for medical and psycho-social support for both former child soldiers and adolescents in general.

78

Martin Sjøen Master Thesis University of Agder

4.7. Summary of main findings part two This analytical part which is based on the first research question, seeks to identify the common socioeconomic traits of former child soldier, when reintegrating back to their local or host communities. The empirical findings outlined in the above sections, demonstrates some common and partly disconcerting features. Firstly, the most prominent challenge faced by all former child soldiers in this sample is related to obtaining adequate economic livelihoods. This implication transcends any social background or situation among the respondents, and for the majority, it persists even years after they have been reintegrated. Further, most former child soldiers believe that their soldiering background is central to this troublesome economic situation. However, there is much uncertainty regarding the statistical relevance of this statement. With the general economic situation in DRC being rather abysmal for the majority of the population, the economic relationship between former soldiers and their non-combatant peers is not known. It is therefore difficult at this stage to conclude what the actual economic impact armed conflict has on former child soldiers. Secondly, the fact that a majority of respondents were able to reunite both with their families/relatives, and subsequently also their local communities, is indubitably an encouraging feature. There is broad consensus concerning this both from the respondents and in the academic field at large (Wessells 2006: 183). These findings show a considerably higher proportion of family reunification than previous studies in DRC (ILO 2003: 55). Still, the majority of respondents described various degrees of reluctance from family and community to socially accept them back. On the basis of the available accounts, prejudice, discrimination and stigmatization were all prevalent among a considerable segment of the interviewees. Anecdotal evidence suggests that community reintegration is significantly more strenuous than family reunification. Still, there are strong indicators that imply an increase in social acceptance over time. For example, former child soldiers who were active around the period of the First Congolese War spoke of less dismissive social relations, and were more financially independent than younger respondents. To some extent, these social trends are comparable to other case studies around the world (Annan et al. 2010). Despite these social shortcomings in the reintegration process, most respondents hold an affirmative approach to their social well-being. There is a common view that sensitization work is a sufficient approach to address these social implications. Some respondents also demonstrate high level of active agency, as they claim considerable “ownership” to becoming socially accepted themselves. A belonging

79

Martin Sjøen Master Thesis University of Agder to social entities like family and community is believably paramount for all the respondents. Thirdly, local communities are more divided in their perception(s) of child soldier reintegration. While there to some extent is a common view regarding the overall responsibility of communities to accept “their children” back, this aspect lies ultimately with the young combatants’ respective families. Furthermore, it is also dependent on the individual background (crimes committed) of the former child soldier. On one hand, this view “defies” legal aspects of reintegrating underage soldiers. Nonetheless, it is probably based on the widespread suspicion and animosity most civilians have towards soldiers in DRC. The overall impressions from many members of local communities render little doubt that their human security is somewhat threatened in the presence of former child soldiers. Fourth point, former child soldiers in this research gave scarcely any impressions of hostile behaviour, consequently renouncing all acts of violence. This finding contradicts the “mainstream” impression of former child soldiers, yet it finds strong support from comparable academic studies (Annan et al. 2010: 6-7). There is however indications that some former child soldiers are more prone or vulnerable to perform hostile acts or behaviour especially in the early stages of their reintegration process (often around the demobilization process). Hostility towards former child soldiers is on the other hand a neglected topic. According to some of the respondents, hostile and violent acts towards former child soldiers may be a significant and underreported subject. It seems evident that this situation impairs the human security of both former child soldiers, and their local communities, as will be discussed in part three of the analytical section. Finally, the appropriateness of Western psychological/medical paradigms in non- Western settings is a current topic. Regardless the immense focus on the mental well-being of children and adolescents in armed conflict in recent years, the contextual circumstance of this aspect in DRC is characterized by ambiguity. There are few studies on the relationship between psychological distress factors and overall mental well-being. Based on available studies, tentative findings suggests that a large part of Congolese child soldiers experience traumatic exposure. Psychosocial assistance towards former child soldiers is somewhat limited. Yet the only significant appeal among former child soldiers was more long-term follow-up support.

80

Martin Sjøen Master Thesis University of Agder

Part 3 4.8. Assessing intervention efforts in the Kivus The foregoing data presents common traits among former child soldiers when reintegrating back to society. Furthermore, these traits illustrate noteworthy socioeconomic challenges, which potentially may hamper a successful socioeconomic reintegration. One should avoid labeling these findings as “universal truths”, yet they are widespread, and to a great extent comparable to case studies from other countries (Blattman and Annan 2010). Child soldier intervention programmes are as mentioned part of recovery strategies encompassing “economic development, security sector reform, integration of refugees [with] justice and reconciliation” (International Peace Academy 2002: 3 in: Schroeder 2004: 3). They are aimed at assisting former child soldiers back to a civilian life. While some studies show that intervention programmes receive “positive overall reviews” from former child soldiers (Boothby 2006, Humphreys and Weinstein 2004), little is known on how they actually impact the reintegration process. Based on the available testimonies, how has formalized intervention programmes impacted the socioeconomic reintegration process for the former child soldiers in question? Moreover, can these measures be characterized as effective conflict transformational efforts in the Kivus? For clarity, to distinguish between former child soldiers who have participated in intervention programmes and those who have not, the former are referred to as “participants”, while the latter as “non-participants”. When describing the entire sample, “respondents” will still be used.

4.8.1. Intervention efforts towards social reintegration All eighteen participants in this research had been actively approached by people affiliated with intervention programmes, or through the direct release of underage soldiers from armed groups. Most participants believed they would not have been recipients of intervention support had they not been actively approached. The six non-participants were not familiar with intervention programmes at the time of demobilization or reintegration, nor had they been offered this support in retrospect. Yet when asked, all six non-participants confirmed they would have preferred receiving reintegration support. The effects of intervention programmes reveal a somewhat diverging perspective between the participants’ perceptions, and the observations made by the author. Among the eighteen participants, the large majority (fifteen) described a sincere gratitude in being recipients of intervention support. By large, these participants valued the recognition, support and training they were given. Yet, the effect of participating in intervention programmes

81

Martin Sjøen Master Thesis University of Agder appears to be tied more with individual attributes, than a strengthening of the collective cohesion. This divide in perceptions versus observations adheres therefore to the gap between the satisfactions described by many participants, against the persisting social challenges they nonetheless experienced. After all, prejudice discrimination and stigmatization was frequently reported by the majority of both participants and non-participants. In fact, there are few indications that participating in intervention programmes has contributed noteworthy in restoring social relations with families or communities. This consistency in demographic profile among respondents is indicative of a serious and persistent social challenge that requires comprehensive attention. Especially within community reintegration, as this process appears to be substantially more challenging than family reintegration. The three respondents who were unable to reunite with their families had all participated in intervention programmes, and they all described unsupportive families pre- recruitment. Clearly the intervention support had not been able to assist them adequately. Moreover, none of the respondents from the focus groups (community members), had received any notable training on how to provide sensitization and reintegration support for former child soldiers. This finding may be arbitrary, and it is difficult to verify its external validity. However, several practitioners in Bukavu claimed that there were few communities in South Kivu who received formalized reintegration support. Apparently this had worsened over the course of the last year as international funding had “dried up”. Local churches conduct most “reconciliation efforts”, and NGOs and CBOs are often unable to perform the immense peacebuilding work that is required in this region. Comparable to other studies, social acceptance seems to increase over time making the former child soldiers “function at par with others in their community” (Annan et al. 2010: 5). Yet most respondents did not ascribe this to formalized intervention efforts per se, but rather their families and surrounding social environments. Interestingly, “family” seems to be a correlative factor for explaining both voluntarily recruitment and successful reintegration. Two of the six non-participants were in actuality among the seven respondents in total who had become more “independent”, eventually establishing their own families and homes. Moreover, participation in intervention programmes is likely not the explanation for why the high number of overall sampled respondents (88%) were able to return to their homes of origin. Non-participants show also high rates of family reunification, and this seems related to the widespread view on the foundation of family belonging, demonstrated both by former child soldiers and members of local communities. What is more, this has comparable features with other studies both in DRC and Africa in general (Blattman 2009, Boothby 2006,

82

Martin Sjøen Master Thesis University of Agder

Humphreys and Weinstein 2004, Verhey 2003). Based on the interviews with former child soldiers, all conveying sincere appeals to be socially accepted, conjectural evidence implies that the reintegration process is only partially dependent on the child soldiers themselves. Contextual variables such as family and community behaviour are as crucial or more, as the actual efforts of formalized intervention and reintegration. Intervention programmes seems largely negligible in the reintegration process for the majority of participants altogether. There is also a visible lack of long-term social follow-up and psycho-social support for participants. While a significant part of the respondents believed that family and community sensitization would be beneficial, anecdotal evidence suggest that it has had seemingly little visible effect on restoring social relations. It should be noted in this context that different intervention programmes have different approaches (non-standardized) with regards to sensitization and follow-up support. Observations made by the author reflect therefore an exterior and general view of several different intervention programmes rather than in-depth study of a specific program. Still, this finding contradict comparable studies which have found sensitization efforts as beneficial in some communities (Boothby 2006: 253, Wessells 2006: 186), thus two important considerations must be made. First, the lack of systematic study of an intervention program and its effect on a specific community may affect the external validity. Second, comparable peacebuilding efforts have up-to-date predominantly been conducted in post-conflict settings. In light of the unstable sociopolitical situation in DRC and the shortfall of effective peacebuilding efforts, a direct comparison may not be suitable. Overall, the findings in this research provide little affirmative knowledge on the social effects of intervention programmes. Yet in this sample, participants are evidently not better socially integrated than non-participants. Sociocultural factors are likely more important than formalized intervention efforts. The importance of traditional soldier – civilian relationships in the context of DRC is another important aspect, and this requires more attention. Further, it seems as if the sociocultural part of intervention programmes suffers from several shortcomings, especially the lack of long-term follow-up support. However, the general lack community sensitization should perhaps be characterized as a shortage “of” intervention programmes rather than a shortcoming “in” them. Simultaneously, many participants voice considerable satisfaction with intervention programmes which should not be trivialized. What is more, none of the participants have had any affiliation with armed groups after departing from their respective outfit. This may indicate that participation in intervention programmes may reduce or even prevent re-recruitment of child soldiers.

83

Martin Sjøen Master Thesis University of Agder

The overall inconclusiveness surrounding intervention efforts supports the argument put forward by Blattman and Annan, namely that the “real effects” of intervention efforts are still largely unknown (2008a: 24). This evidence suggests that intervention programmes have insignificant effect on social reintegration, and that contextual factors such as poverty, culture and the ongoing conflict are perhaps the most important factors in determining a successful socioeconomic reintegration.

4.8.2. Intervention efforts towards education and economic restoration The economic profile of former child soldiers in the Kivus is quite similar across factions. Based on the evidence, acquiring a sustainable livelihood is the most prominent challenge for (male) former child soldiers. Accordingly, this is consistent with dominant theories on child soldier reintegration (Blattman and Annan 2008a, Boothby 2006, Humphreys and Weinstein 2004). Furthermore, this seems accepted by practitioners in the field as most intervention programmes focus extensively on building economic capacity through educational efforts and vocational training. Every participant interviewed had been offered vocational training, education or material benefits for facilitating income generating work. Conversely, the effect of economic intervention should perhaps be markedly more apparent than social intervention efforts. In that respect, what are the effects of educational and economic support and to what extent have these measures been effective in finding or creating meaningful occupations for the participants? First and foremost, the majority of participants showed great appreciation for having been given educational and vocational support. Equal to that of receiving social support, this assistance provided the former child soldiers’ with a much needed recognition. More than anything, it appeared as if participants were given a larger sense of purpose, and it allowed them to pursue individual goals. With most participants finding their initial educational support helpful, the long-term outcomes were less affirmative. Evidence suggests that intervention programmes are not tailored adequately according to the individual needs of the participant. For example, few participants had achieved ordinary level education pre-recruitment, yet only two of eighteen chose to complete their basic education after being demobilized. None of the participants had been offered customized basic/higher education based on their specific age or individual background. When being asked, most participants confirmed that they would like to acquire skills such as reading and writing. Still, they did not assess the possibility of embarking on basic education as a realistic option due to certain barriers. Firstly, the participants described

84

Martin Sjøen Master Thesis University of Agder understandably little enthusiasm on the potential of being put in the same classroom as children perhaps half their age or younger. Many of the respondents considered themselves perhaps rightly as adults.

One respondent illustrates this dilemma:

In the Mayi-Mayi, I was the commander of many soldiers. I had the responsibility of training new recruits. Now these people ask me if I want to go to school with small children.

Another challenge refers to the limited support towards education many participants were offered. With education being a costly affair for most Congolese, the prospect of commencing a study program one may not be able to complete did not appeal to most participants. Evidence from comparable studies suggests that schooling loss creates education gaps which may affect earning gaps (Blattman and Annan 2008a: 16). The majority of participants chose rather vocational training as their educational intervention strategy. Male former child soldiers may traditionally select from professions as brick making/laying, carpentry, tailoring, cultivator, mechanic, or other practical vocations. Generally this involves a time-limited training followed by being given a “start kit” or toolbox designed for the specific profession. In this survey, most participants had chosen training within carpentry or brick making/laying. Nonetheless, most respondents were currently engaged in cultivation, mainly family-based economy, as this was seen as a better guarantee for providing food. Support for finding adequate work seemed absent and this was the basis for the most common complaint. On reflection, several participants described their vocational training background as futile considering how they felt educated towards unemployment. There were seemingly little difference between participants and non-participants regarding the difficulty in acquiring a sustainable livelihood. According to Dallaire, intervention efforts must carefully consider the vocational support given so that it matches productivity needs that actually exists in the community (Dallaire 2011: 168). For Congolese men, this is especially important as they are expected to provide for their families financially. To the author’s impression, this is not sufficiently considered as vocational options are limited to only a handful often “saturated” professions. Senior researcher Kåre Lode described his experience during a study of child soldier intervention work. In a village in Ivory Coast, an excess of seventy former child soldiers had received vocational training to become mechanics in a community without any motorized

85

Martin Sjøen Master Thesis University of Agder vehicles18. Such a case is perhaps among the more extreme examples; yet this troublesome situation has comparable features to intervention work being carried out in the Kivus. Several of the participants expressed considerable disappointment when explaining how they had abandoned the intended vocation they had trained for. There is however a tendency that vocational training may have a positive effect on a very specific part of the social reintegration process, namely family reunification. Former child soldiers may be perceived as more useful by their relatives because of their newly acquired skills. Practical rewards such as start kits may also become beneficial for the family, sometimes even sold to generate income according to the participants. On the other hand, this type of material benefits may have a counterproductive effect on local communities, as they watch “violent perpetrators” being financially rewarded. Many participants described how the affiliation of intervention programmes could reinforce derogatory views. Several authors have emphasized the prospective challenge of awarding former child soldiers financially (Dallaire 2011: 161, Wessells 2006: 180). Besides the already mentioned potential for creating additional social conflict between former child soldiers and their local communities, it also reinforces the divide between former child soldiers and their non-combatant peers. Moreover, it may become an incentive for children and adolescents to lie about their background in order to receive financial rewards. According to the evidence presented, the effects of intervention work on education and vocational training is a complex matter. While it may provide former child soldiers with a sense of individual enrichment and beneficial skills, the overall advantage is rather unclear. As described earlier, the economic situation in the Kivus is challenging. Speculative evidence suggests that former child soldiers may be relatively worse off than non-combatant civilians. Yet the potential gap between combatants and non-combatants needs to be further explored. While poverty seems endemic in most parts of a war-torn DRC, it is perhaps not realistic to assume that intervention programmes are sufficient measures to address economic challenges in isolation. Participants should presumably gain benefits from educational and economic support. Considering the testimonies presented here this is not the case. Despite the various formalized efforts towards reintegration, the economic consequence of being a former child soldier is substantial. This corresponds with leading theories (Blattman and Annan 2008a). Furthermore, intervention efforts are largely structured according to a “one-size-fits-all”

18 Lode, K. Senior Researcher at Centre for Intercultural Communication (SIK) Lecture in Stavanger 14.03.2012. 86

Martin Sjøen Master Thesis University of Agder strategy, neglecting the individual preferences of former child soldiers. It is also questionable as to whether participants in this sample have been given training that may actually be utilized in their respective communities.

4.8.3. Contextual dimensions towards intervention work The foregoing data on the effects of intervention work may give rise to despair. Evidence presented highlight considerable socioeconomic challenges for most respondents. Economic issues are especially prominent, and there are few indications that intervention efforts alone are sufficiently addressing these challenges. From a Western perspective in particular, this is problematic. Despite the “desolated” testimonies on intervention efforts, most respondents in this research show strong resilient behaviour. For example, while many reported prejudice, discrimination and stigmatization, nearly two-thirds of the respondents described an increase in social acceptance over time. “Only” two out of the twenty-one respondents who reunited with their families expressed considerable dissatisfaction with their present living situation. While these features challenge some of the few empirical data that exists from DRC (ILO 2003), they confirm findings from other countries (Annan et al. 2010, Blattman 2009, Boothby 2006). What is more, these findings ask the question whether it is suitable to assess intervention efforts in isolation, or if child soldier reintegration should be viewed in a larger and more culturally appropriate context. In accordance with a stated methodological objective, participatory research will be emphasized in this research. Some authors argue the need for current research to explore the circumstances of child soldiering from the perspectives of the young combatants themselves (Hart and Tyrer 2006, Tonheim 2010). As shown in the upcoming diagram, the majority of participants expressed considerable satisfaction with their intervention support. While many former child soldiers are faced with a series of socioeconomic challenges, the high rates of satisfied participants, subsequently sharing features with other studies (Boothby 2006, Humphreys and Weinstein 2004), should not be considered as a negligible detail. Personal development as experienced and defined by the participants must be incorporated when assessing the “success” of intervention efforts. Keeping in mind the philosophic perception on intervention efforts (the maximalist perspective), they are best viewed as “a temporary process aimed at establishing the preconditions for development” (Muggah et al. 2003: 5). One can certainly argue whether these intervention efforts have established sufficient preconditions for development, as most

87

Martin Sjøen Master Thesis University of Agder participants are troubled with lack of obtaining sustainable livelihood. Yet the common perception of gratitude expressed by participants should at least be viewed as a contextual advantage.

Perception of intervention efforts

17%

Satisfied

83% Unsatisfied

Figure 15: Diagram on the perception of intervention efforts. Furthermore, some of these views assert primarily collective well-being before the individual. For example, almost one-quarter of the participants described how they were the biggest guarantee for a successful reintegration themselves. Put in other words, they had to prove to their surrounding social setting that they truly had “good” intentions. Participants also demonstrate reassuring views on their current situation and future prospects. There seems to be little doubt that such perceptions eschew the “individual perspective” often found in leading Western paradigms (Boyden 2010: Preface). Program director for the German NGO GIZ in eastern DRC Achim Koch, believes that child soldier reintegration is less “problematic” than we often perceive it to be. According to Koch, the culture of the region will often be the most decisive element regarding a successful reintegration as he claims that:

[Child soldiers] will manage their life as all others do … it seems that the Congolese people are not very interested in the past19

In question whether intervention efforts as studied can be characterized as effective measures, an answer would largely be case dependent, conditional on the respective analytical

19 Achim, K. Program Director at the Die Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ). E-Mail: 16.01.2012. 88

Martin Sjøen Master Thesis University of Agder perspective. While there is much ambiguity regarding the socioeconomic impact of intervention efforts, in fact there are few indications that they contribute noteworthy in restoring social relations or establish sustainable livelihoods. Participants still express satisfaction in having partaken in such programmes. There are in other words contradicting views on these efforts. At the same time, there is reason to assert these opposing views to derive from different sociocultural perspectives between the researcher and the respondents. The former who is firmly based within a liberal Western paradigm view the reintegration process mainly from an “individual” (micro-level) perspective. The latter however, tend to proclaim a more “collective” (macro-level) perspective. According to Boyden and de Berry, as armed conflict tends to impair most aspect of society, a strict focus on the “individual” may result in losing important dimensions of the conflict (2007: xiv). What is more, tentative findings suggests that there is basis to question the effectiveness of general peacebuilding efforts in eastern DRC (Autesserre 2007: 438, Dagne 2011: 5, Tonheim 2011: 3-4). Expectations that child soldier intervention programmes should “transcend” these challenges may not be a realistic. Overall, these aspects make conclusive attempts rather assumption based and should therefore be cautiously applied. The effectiveness of intervention programmes towards the general peacebuilding efforts in the Kivus will further be explored in section 4.9.1.

4.8.4. Key patterns of intervention efforts Evidence presented here does not derive from an in-debt study of a specific intervention program in the Kivus. These findings are largely un-generalizable due to none- representative sampling and few respondents. The anecdotal findings are therefore best viewed as a contextual study of the “lessons learned”. Thus it serves as an incentive for conducting a more rigorous investigation of the phenomenon in question. However, there are many commonalities among the testimonies of the respondents; hence key patterns can be drawn from them.

 The vast majority of participants’ express sincere gratitude for the support they have been given. By large, this gratitude transcends most challenges or shortcomings towards the respective reintegration process. Intervention efforts may be an important factor in providing much-needed recognition and creating awareness for participants.

 Based on how most participants were actively approached by staff from intervention programmes, evidence would suggest that these efforts are conducting comprehensive

89

Martin Sjøen Master Thesis University of Agder

work to identify, support and incorporate former child soldiers in their programmes. Furthermore, the existing legislative framework in DRC seems imperative for the release of former child soldiers by armed groups, albeit recruitment still being a challenge in regional parts of the country. Available testimonies indicate that intervention programmes may further prevent or reduce re-recruitment.

 Educational support suffers from not being contextually adapted to the individual participants while also being temporal limited. Subsequently, this implication appears to discourage many participants from embarking on basic education against their preference.

 Vocational training seems to have positive initial effect, yet less affirmative long-term effect. Many participants experience being skilled for a profession which they are unable to obtain after training. Moreover, long-term support to find sustainable livelihood seems largely absent.

 Depending on the vocational training and material benefits, intervention support may positively impact a former child soldiers’ family reunification. On the other hand, such support may further create social conflict in local communities.

 Hostile behaviour towards soldiers in DRC seems extensive, and conversely, this social inclination should be understood in a historic perspective. While hostility for soldiers in general illustrates the importance of conducting community sensitization, the apparent lack of involving stakeholders in intervention programmes may reinforce the widespread derogatory view towards former child soldiers.

 Intervention programmes do not adequately target simultaneously the former child soldiers and his family. At the same time, sociocultural aspects and especially the participants' family seems to be perhaps the main causal factors for a successful reintegration. Former child soldiers appear to be less affected by socioeconomic challenges when being adequately supported by immediate or extended family.

 Specialized psycho-social and medical support appear largely absent among former child soldiers. Anecdotal evidence suggests that child soldiers in DRC experience a significant amount of traumatic events, thus it deserves arguably a greater focus.

90

Martin Sjøen Master Thesis University of Agder

4.9. Peacebuilding through conflict transformation The following section will assess the effects of formalized child soldier intervention efforts through conflict transformation theories. Although peace and conflict studies is a young discipline, scholars and practitioners have gradually constructed the philosophy of conflict transformation based on prevailing theories within the field (Deutsch et al. 2006) 20. An important concept which has emerged from this interdisciplinary field involves the differentiation between destructive conflict, and constructive conflict respectively. In this analytical section, the former encompass mainly the armed conflict in DRC, alongside detrimental elements such as hostile relationships, interests and discourses. The latter however, will involve the process of transforming destructive into constructive conflict (Miall 2004: 4). According to Galtung, conflict is inevitable and has both life-affirming and life- destroying elements. A transformation from one to the other occurs gradually through a series of smaller steps, often involving several actors (1996: 90 in: Miall 2004: 4). Figure 16: Course of conflict.

Destructive conflict: Constructive conflict: Behaviour which transforms Behaviour which sustains or destructive conflict through augments the conflict cooperation

Key aspects: Key aspects: - Suppressive actors - Accommodating actors - Low social acceptance - Social recognition - Security needs unmet - Broad human security - Exploitative economic - Supportive economic structures structures - Illegitimate government - Legitimate government

Conflict escalates Conflict de-escalated

Figure 16 illustrates part of Azar’s (1990) framework for transforming social protracted conflicts.

20 Conflict- resolution and transformation are interchangeably used in literature. Deutsch et al. (2006) prefers conflict resolution in their work. Yet the prerequisites they “apply” this concept, makes it somewhat consistent with the understanding of conflict transformation as defined by Miall (2004) and Ramsbotham et al. (2008). 91

Martin Sjøen Master Thesis University of Agder

It may seem contradicting that a particular conflict can contain elements from both destructive- and constructive conflict respectively. However, this is believably tied with dynamics of armed conflict. Conflict transformation is rarely viewed as a linear process, meaning that the sequencing of transformational changes may vary depending on the “logic of the situation” (Miall 2004: 11). In other words, certain aspects, actors or regions may experience a more rapid or immediate conflict transformation process than others.

4.9.1. Destructive conflict It seems evident that the process of social reconciliation suffers peculiarly as the majority of respondents reported strained family and community relations. Nearly half the respondents believe they were regularly perceived and/or treated differently by their families, and a total of 80% had experienced recent acts of prejudice, discrimination or stigmatization within their communities. Deutsch et al. (2006: 10) define these challenges as interpersonal- and intergroup conflicts, and argue that they are often prolonged and intense in nature. According to the respondents, this treatment is due to the characterization of being a former soldier, which may correlate with the “derogatory” view many civilians carry towards members of armed forces in DRC (Baaz and Stern 2008). There is seemingly no difference between participants and non-participants on this matter. Poor relationships between actors have the potential of further escalating conflicts, and it remains a critical hindrance to peacebuilding efforts (Miall 2004: 8). All participants interviewed in this research have stayed in their current homes for at least one year, which may imply that the reintegration process is beyond the “inceptive” stage. These social implications seem therefore both in-depth and sustained. Does this indicate that the social conflict relating to child soldier reintegration is not “transformed” in its deepest sense? With the disparaging views on Congolese soldiers in general, evidence may imply that the social aspect within conflict transformation theory is severely damaged and perhaps even non-existent (Miall 2004: 8). According to some theories, this situation suggests a destructive nature of conflict (Deutsch 1973). Furthermore, there are few signs of inclusive efforts of peacebuilding through the transforming of relationships between former child soldiers and members of society. Intervention programmes seems unable to reach any significant number of local communities in what should be awareness building. Thus it clashes with the dominant approach which proposes peacebuilding at the grassroots level (Rupesinghe 1995, 1998 in Miall 2004). In actuality, the absence of involving civil society in peacebuilding efforts may be seen as a

92

Martin Sjøen Master Thesis University of Agder partial explanation for the sustainment of the conflict in DRC. This degenerating situation is further reinforced by the fact that most participants are unable to obtain sustainable livelihoods. Conflict on both micro and macro level is believed to be greatly influenced by economic factors. For example, recent efforts in West Africa have shown the “futility” in attempting conflict transformation without addressing economic interests (Miall 2004: 9). The correlation between poverty and armed conflict has been thoroughly documented in recent years (Blattman and Miguel 2010, Collier 2008, UNDP 2009). Poor countries are both more exposed to violent conflict, but the conflict itself will inflict further financial costs on its population, making the situation more intractable. Based on the available evidence, conflict transformation through intervention work seems to have had little visible impact on facilitating economic growth for the former child soldiers. Undoubtedly a limitation as this factor exacerbates the most prominent challenge for most former child soldiers. Although this research lacks sufficient data to comprise an adequate analysis of the human security for the affected communities, the exploratory evidence suggests that neither former child soldiers, nor other stakeholders in the Kivus receive adequate protection from impending threats (human security). Hostile behaviour towards former child soldiers seems extensive, and many respondents describe being the victims of unprovoked violence. Conversely, respondents from local communities depict a situation where they clearly perceive their human security as compromised. A common argument among local communities is that former child soldiers exhibit violent and demeanour behaviour. Moreover, many community members claim that former child soldiers cannot be trusted, and that drug use and criminal behaviour occurs frequently among them. The realities of these claims are questionable, and to a large extent they contrasts the views described by former child soldiers. Comparable studies from different countries show that the majority of former child soldiers show modest or no aggression in post-conflict settings (Blattman and Annan 2010, Boothby 2006, Humphreys and Weinstein 2004). Yet these perceptions must be taken into consideration as fear and hostility form a vital aspect in the conflict transformation framework. These findings may imply that the intervention efforts are unable to transform the relationship that “supports the continuation of violent conflict” (Miall 2004: 4).

While the external validity of these perspectives is difficult to verify, they show how former child soldiers and local communities are truly not “free from fear”. By large, this destructive perspective is augmented when controlling with available data, as findings from

93

Martin Sjøen Master Thesis University of Agder different studies confirm that a significant proportion of the population have experienced severe traumatic exposure (Bayer et al. 2007, ILO 2003, Johnson et al. 2010, Mels et al. 2009). If supplemented with the recent Human Development Index, it seems evident that the prevailing poor human development deprives most people in the region from their “freedom from want” (UNDP 2011).

4.9.2. Consequence of destructive conflict Building on the previous section, the transformational effects of intervention programmes in the Kivus are somewhat unclear. Evidently, security needs of both former child soldiers and local communities, which is a precondition for development (McIntyre and Weiss 2003), are not adequately addressed. The socioeconomic situation for most former child soldiers may create a degenerating and negative spiral, and the nature of this situation has clear characteristics of a destructive conflict. A destructive nature of conflict has further the consequence of increasing the risk of prolonging the armed conflict. Collier, who refers to this as the “conflict trap”, claims that the cost of a major conflict is immense (2008). To a great extent, Miall concurs as he argues that destructive conflicts intensify the damages towards the “system of governance, the economic order and the social institution of society” (2004: 12). Communities that are characterized by destructive conflicts are further disadvantaged by “losing” their capacity to transform conflicts, subsequently making peacebuilding efforts like intervention programmes even more strenuous (Miall 2004: 11).

4.9.3. Constructive conflict transformation An exclusive assessment on the effects of intervention programmes in the Kivus is arguably a narrow focus as it excludes the greater context within where the conflict is situated. According to Miall, if conflict is restrictedly viewed, contextual aspects required to understand the conflict may be lost (2004). Although a study on the effects of child soldier intervention efforts is both needed in general, and specific to this work, it nonetheless “detaches” this process from the overall peacebuilding efforts in the Kivus. Thus a broader approach within the child soldier reintegration process should be included. Here follows four important aspects which views intervention efforts in the Kivus within a wider perspective. First aspect includes the social improvement described by nearly two-thirds of the respondents, illustrated in the diagram below. There is perhaps little reason to assert this self- reported social improvement solely to intervention efforts being carried out in the Kivus. Vocational training may contribute towards family reunification in a positive way, yet there are few indications that it offers other benefits. Particularly seeing that respondents from local

94

Martin Sjøen Master Thesis University of Agder communities describe little experience of adequate reintegration support, and none of the participants received long-term social follow-up. Yet the sequencing of positive social change and reassuring views expressed here carries in fact important features from conflict transformation theory. The element of altering relationships in a positive way lay namely at the heart of this peacebuilding philosophy (Ramsbotham et al. 2008: 29).

Increasing social acceptance

Yes No Community

Yes No Family

0 % 20 % 40 % 60 % 80 % 100 %

Figure 17: Diagram on social acceptance. Restoring or transforming social relations is unquestionably a complex process, and in some cases it may not be a linear one. Miall further states that only in the very “simplest conflicts is conflict transformation likely to be a rapid or immediate process” (2004: 11). He also claims that the process is usually slow and tortuous, with evaluation of it being extremely difficult. Such a description fits well with available testimonies, and parallels can be drawn from the philosophy of conflict transformation, to the reintegration process studied in this context. Empirical evidence implies also that participation in intervention programmes could reduce or even prevent re-recruitment. While there are few reliable statistics, data from the two Congolese Wars suggests that nearly one-third of all combatants who have been involved in the fighting are children (Achvarina and Reich 2010: 59). A figure which is considerably higher than the estimated 10% of children involved in armed groups around the world (Singer 2006: 30). Addressing child soldier recruitment should therefore be considered as part of conflict transformation.

95

Martin Sjøen Master Thesis University of Agder

Secondly, albeit being fairly inconsistent among respondents, there is evidence of what Mitchell characterizes as transformation on the local and individual level (2002). For example, the majority of community members claimed they acknowledged the reintegration of former child soldiers depending on the “will” of the family. A sense of responsibility was also present with one-quarter of the interviewed former child soldiers claiming they were themselves responsible for becoming socially accepted by their families. Participants show also an understanding for how relatives could view them “differently”. Examples of personal changes like these are argued to be crucial. While it may be difficult to assess precisely how local actors through micro level conflict management impact the overall peacebuilding process, it is nonetheless believed that this transformation can have an affirmative effect on personal and group dynamics (Miall 2004: 10,14). Moreover, ending armed conflict through “peaceful negotiation” on macro level has had negligible effect in contemporary warfare (Miall 2004: 16). Perhaps this implies that social protracted conflicts are best addressed through a constructive change in social attitudes. Thirdly, conflict transformation as defined requires a wide range of actors to participate, with local communities being the most important (Miall 2004: 14). It may therefore not be fitting to expect NGOs to be the main transformational actors. The change must ultimately come from within the affected communities (civil society), as formalized intervention efforts are believed to have a limited effect on transforming the underlying conflict (Miall 2004: 12). According to Ramsbotham et al. (2008: 198), the process of structural peacebuilding usually commence well after intervention efforts, which are believed to occur in the early phase. Findings in this research may imply that different actors, especially family and relatives contribute predominantly towards the described social improvement. Anecdotal evidence suggests also that institutions such as local churches conduct considerably towards the peacebuilding process. Thus it provides strong evidence that “people within the society” are transforming social conflicts through a gradual peacebuilding work (Miall 2004: 4). Lastly, the evidence to support the claim that former child soldiers are economically more disadvantaged than other Congolese is weak at best. The findings in this research are arguably too tentative to generalize upon. With DRC being one of the poorest and least development countries in the world (UNDP 2011), there are few reliable data indicating that former child soldiers are significantly worse off than other vulnerable groups or the population in general. Traditionally, most Congolese are dependent on the informal sector to provide food and other goods on a daily basis (Koyame and Clark 2002). Based on these

96

Martin Sjøen Master Thesis University of Agder testimonies, the “plight” of former child soldiers in question seems not to differ much from the population at large. Intervention efforts may under right circumstance facilitate income- earning work for former child soldiers, but this is very challenging while the underlying economic situation in DRC remains as it is. In fact, contemporary conflict resolution theories often place the economic (re)-construction within the post-war phase (Ramsbotham et al. 2008: 207). There is little evidence supporting the claim that northeastern DRC should be categorized as a post-conflict region. What is more, experiences from countries like DRC suggest that the “legacies of war economies” pose significant problems to formalized intervention efforts. The chief focus in these states is argued should be the withdrawal of violence out of the economy, a task Ballentine and Nitzschke claim is “easier said than done” (2005: 15). It should therefore be critically assessed what the realistic economic expectations of intervention efforts should be in the Kivus.

4.9.4. Consequence of constructive conflict transformation The transformations described in the previous section are not the sole effects of intervention programmes. Various sociocultural factors are likely more important elements in the larger peacebuilding process. To analyze intervention efforts within conflict transformation theories requires therefore a broader consideration. Should these aspects be analyzed within the theory of conflict transformation, one can additionally speculate the further consequence of peacebuilding in the Kivus. An effect of this transformation would likely be the further rebuilding of social relationships. The general peacebuilding is argued to be “futile” without such transformation (Mitchell 2002). There are clear signs of its presence, mainly in the significant improvement of social acceptance most former child soldiers’ experience. Moreover, “older” respondents describe less socioeconomic challenges, and appear to be better reintegrated than their younger peers. It is therefore likely that improved social structures will function as a stabilizer in this conflict prone region. Peacebuilding through conflict transformation theories are generally expected to be very time consuming in war-torn states like DRC (Miall 2004). In this study, “older” respondents seem better socially reintegrated than their younger peers, and in some cases, better placed financially. The link between social reintegration and income earning is strong. Yet this noticeable link between social- and economic capital refers to a correlation, as the causality here is not appropriately assessed. It may also be the case that economic development improves social relations. Regardless, the consequence of improved social relations will in all probability contribute towards a strengthening of the economic

97

Martin Sjøen Master Thesis University of Agder situation (Annan et al. 2010). Another probable consequence of conflict transformation through intervention efforts would be the likely reduction of child soldiers’ usage/recruitment. The decrease in re- recruitment found in this study may be justified by the increased awareness of former child soldiers through intervention efforts. Subsequently, it may also find its causal explanation in the global efforts to criminalize child soldier usage, or a combination of both factors.

4.9.5. Summary of main findings part three Child soldier intervention (DDR) programmes are generally considered to be among the most recognized post-conflict efforts globally (Muggah et al. 2003). There are many who advocate intervention efforts as potent peacebuilding tools, yet few are able to show what long-term effects they actually possess (Blattman and Annan 2008a: 24). To a great extent, findings from this research mirror this ambiguity. While the majority of participants describe great satisfaction in having participated, there is in fact little evidence to support the claim that they are sufficiently facilitating a successful reintegration. Intervention efforts have seemingly little visible effect as the majority of respondents describe considerable challenges towards the social reintegration. Community reintegration appears to be relatively more strenuous than family reunification. There is seemingly no difference between participants and non-participants. According to community members, reintegration support is not adequately addressed as sensitization work is largely absent in this region. Comparable to other studies, social acceptance seems to increase over time. This improvement should perhaps be understood more within a contextual perspective in DRC. According to the respondents, economic challenges remain the most prominent obstacle towards successful reintegration. For the benefit of former child soldiers, most intervention programmes give great attention towards vocational training. Yet the long-term effect of this focus is less affirmative. Participants are unable to secure sustainable livelihoods after training. What is more, despite aspiration to engage in education, most participants refrain due to lack of individual adaptation. By large, one of the greatest implications towards intervention support is the “one-size-fits-all” approach currently used. The key pattern of intervention efforts shows a mixture of both positive and negative elements. It is difficult to draw a conclusion on the “successfulness” as it depend very much on the position (individual versus group aspect), as well as whether they are viewed in a long- or short term perspective. Besides the mentioned shortcoming, most participants describe great satisfaction in having received intervention support. Non-participants express

98

Martin Sjøen Master Thesis University of Agder furthermore a desire to receive reintegration support. There are numerous of factors including but not limited to the armed conflict, endemic poverty and political turmoil, which may seriously hamper any conflict transformation effort in DRC. It is not likely that intervention efforts in isolation will transcend these factors, and it is important to keep in mind the realistic limitation towards these programmes (Miall 2004: 11). The role of intervention programmes as transformational efforts within the peacebuilding process is another complex matter. A fundamental question remains as to what are the realistic expectations from such intervention efforts? When defined according to the two leading perspectives, “destructive” and “constructive” conflict respectively; it seems evident that the reintegration process of former child soldiers bears features from both. When assessing the challenges faced by most participants as they describe themselves, it would seem likely that intervention efforts contribute limited towards addressing them. Based on testimonies, there is an apparent lack of human security for both former child soldiers and local communities. Lack of effective transformation efforts may further increase the risk of a destructive conflict, in turn undermining the peacebuilding efforts. Regions characterized by destructive conflicts tend to be prolonged, and are more likely to experience a relapsing of violent conflict. On the other hand, if the reintegration process is viewed in a larger context than merely intervention efforts, it carries several features of conflict transformation respectively. Most notably is the transformation on an individual level, which ultimately may lead to positive social change. Based on leading paradigms, using a holistic perspective on conflict transformation is imperative. Overall, there is a recognizable limitation in viewing child soldiering in isolation of the armed conflict in DRC. With child soldiering being a consequence of violent conflict, this detachment may yield a view lacking very important elements. The role of the Congolese government, the economic situation, numerous armed groups and neighboring countries are central components which does not get sufficient attention in this thesis. Furthermore, it is perhaps unlikely to suggest that the human security of local communities will be addressed purely through the demobilization and reintegration of one segment of the armed combatants in DRC, namely the child soldiers.

99

Martin Sjøen Master Thesis University of Agder

Chapter 5: Conclusion The consequential impact of armed conflict on children and adolescents remains one of the most emotionally-driven issues globally. In contemporary warfare, this group is believed to suffer the greatest in terms of injuries, disabilities and casualties. Yet they are not only affected by armed conflict as innocent civilians, with contextual variations, they constitute a segment of the perpetrators of violence. Commonly referred to as “child soldiers”, underage combatants are conceivably involved in most protracted social conflicts around the world. Scholars disagree on the historic perspective of child soldiering. The phenomenon has nonetheless become one of the most important humanitarian topics of the last two decades. In general, the use of children in armed conflict is surrounded by ambiguity, and knowledge gaps are prominent. With few novel exceptions, academic research exploring child soldiering from the perspectives of the young combatants themselves is almost nonexistent. Lack of veritable knowledge may potentially hamper attempts to address this pressing issue in a successful way. The protracted social conflict that has plagued DRC for nearly two decades illustrates a current example of children affected both as perpetrators and victims of violent conflict. Perhaps unprecedented in modern history, the Congolese warfare has left millions injured, displaced and deceased. Child soldiering has periodically been rampant in DRC, with armed groups from more than ten countries employing underage combatants in their violent struggle. Despite a nationwide reduction in child soldier usage, recruitment persists especially in the northeastern parts of the country. At the same time, very little research has been conducted in this conflict-prone region, and relevant information particularly on social reintegration issues is rather assumption based. This work sought to address notable research gaps, specifically in the process of exploring socioeconomic child soldier reintegration within the Kivus, DRC. Special emphasis was put on investigating the effects of formalized intervention work for the former child soldiers, including assessing how these measures affect peacebuilding efforts in this region. By investigating intervention efforts, a greater understanding can be made on contextual dimensions of child soldier recruitment, usage, release and reintegration. Moreover, this knowledge may be central in preventing further (re)-recruitment, whilst also being an important contribution towards facilitating a successful socioeconomic reintegration.

5.1. Contextual child soldier recruitment, usage and departure The sampled unit of former child soldiers provides a brief insight into some of their commonalities. Looking at circumstances which may elucidate child soldiering, it is obvious

100

Martin Sjøen Master Thesis University of Agder that these commonalities both contradicts and confirms available research from DRC and comparable countries. However, testimonies from former child soldiers illustrate also the contextual nature of this field; hence generalization must be cautiously used. Child soldiering in this research is a youth issue proper. With most respondents being recruited in their teens, a reorientation on how people perceive this particular group may therefore be required. In addition, specific sociocultural aspects should further be included in the overall equation. This does not trivialize the use of child soldiers, as it is arguably wrong both legally and morally. Yet the general notion of child soldiers as exploited and traumatized “children” is often a misrepresented perspective, as testimonies reveal that the majority regard themselves as “adults”. There are several accumulating factors (poverty, displacement, family) which may lead to the recruitment of child soldiers. Empirical evidence in this research however, reveals that most child soldiers claim they were coercively recruited by armed groups. The main driving factor is therefore placed within the respective armed groups, and their deliberate policies. To a great extent, the coercive nature of child soldier recruitment is contextually substantiated by the large number of respondents who escaped from the armed groups. Anecdotal evidence suggests that a significant portion of child soldiers are trained and experience active combat. Armed groups deploy child soldiers according to their physical and cognitive development, alongside the respective needs within the military structure. Ongoing debates on what roles and duties are commonly performed by most child soldiers should contemplate this dynamic aspect. Re-recruitment rates are convincingly high, but there have been few creditable efforts in explaining why some are being recruited several times, while most children and adolescents have no such experience. The analytical section involving rudimentary features (demographics) of child soldier recruitment, usage and release, holds a subsidiary role in this research. Yet it has contributed to a further emphasis on a central aspect, namely the role of the armed groups. Although this research lacks comprehensive assessment of this particular detail, it seems evident that armed groups constitute the most important component affecting child soldiering. Contextual factors like poverty, displacement and family should not be underestimated, as they are in various degrees correlative with child soldiering. They are especially imperative in the reintegration process. Still, the key to understanding child soldiering lies foremost within the policies of armed group, as warfare is indisputably a result of more or less deliberate choices. Hence, approaches that aims to address or prevent child soldier recruitment and usage, should contemplate first and foremost the “demand” of child soldiers.

101

Martin Sjøen Master Thesis University of Agder

5.2. Conjecturing common traits of former child soldiers Prevalent traits of former child soldiers are recognizable within this sample, and albeit there being circumstantial variations, the commonalities are prominent. Most child soldiers show resilient behaviour. By large, they hold affirmative outlooks on both current and forthcoming situation, and some demonstrate “active agency”. Despite this, the majority of former child soldiers are implicated by significant socioeconomic challenges, which may potentially hamper their attempts to successfully reintegrate back with their families and local or host communities. According to the testimonies of former child soldiers, the most acute trait is related to the possibility of obtaining a livelihood. The majority is not sufficiently able to be financially independent, and is therefore reliant on families for additional support. This is a severe confinement as men in DRC are expected to not only provide for themselves, but also take care of their immediate and extended families financially. Comparable studies have demonstrated the “futility” in reintegrating former combatants without addressing these economic issues. Thus it is indicative of a serious challenge. At the same time, the overall economic situation in DRC is in an abysmal state. In that respect, there is little evidence to support the claim that former child soldiers are more financially disadvantaged than their non- combatant peers. Further research is required on this particular topic both in DRC and within the field altogether. The legacy of being a former child soldier is socially constrained by the widespread derogatory views in DRC. These perceptions which materialize itself through acts of prejudice, discrimination and stigmatization, is commonly found in local communities and families. Undoubtedly a challenge, as the social setting is vital for the well-being of former child soldiers, subsequently playing a crucial role in their reintegration process. On the other hand, most former child soldiers reunite with their families and local or host communities. Socially, the situation seems to improve over time. What is more, improved social conditions are believed to further strengthen the possibility of obtaining sufficient livelihoods, as these domains are strongly interlinked. The analytical section involving common traits has contributed towards highlighting the importance of contextually appropriate assessments. The lessons learned involves the “acknowledgement” of sociocultural aspects such as group belonging. In fact, evidence implies that factors such as family and social roles are the most decisive aspect within the reintegration process. Perhaps then, reintegration should be considered more through contextually appropriate perspectives, and less through predetermined Western perception.

102

Martin Sjøen Master Thesis University of Agder

5.3. Reasoning the effects of intervention efforts Formalized intervention efforts remains one of the most recognized instruments for demilitarizing and reintegrating former child soldiers back to their families and communities. Thus it is seen as a prominent peacebuilding tool. Yet few scholars have been able to reliably assess the “actual” effects of formalized intervention efforts. This research differs not markedly from that trend. In fact, it is difficult to conclude whether these efforts carry any noteworthy socioeconomic impacts. This is vital to contemplate, especially when considering the “implicated” traits of former child soldiers. There is strong consistency between former child soldiers who have participated in intervention programmes, and those who have not. These testimonies are indicative of a noticeable shortfall, as intervention efforts are presumptively expected to provide notable socioeconomic support for participants. Yet long-term attributive effects of intervention programmes seem absent, and the overall contribution towards the peacebuilding process is unclear. In spite of that, participants of intervention programmes express considerable satisfaction with their affiliation. Thus it can be seen advantageous for the participants, albeit perhaps not in direct relation to reintegration “needs” as traditionally defined. The satisfaction described by participants is not negligible as it provides the individual with a much needed recognition and self-awareness. Social relationships improve over time for the majority of child soldiers, demonstrating affirmative change. Established peacebuilding theories emphasize the feature of conflict transformation on an interpersonal level. Furthermore, the protracted social conflicts in DRC have embraced most aspects of society, and it is debatable whether intervention efforts are expected to transcend these destructive consequences in isolation. This thesis found that when these points are examined together (they are seemingly contradictory), it becomes clear that structural peacebuilding is an intricate and long-lasting process. Additionally, these testimonies invite further analysis of leading perceptions (individual versus group perspectives), and argues that the peacebuilding efforts requires more holistic and contextually appropriate intervention efforts. Further research is needed when addressing knowledge gaps and improving the current practice of child soldier reintegration.

103

Martin Sjøen Master Thesis University of Agder

References Achvarina, V., Nordås, R., Østby, G. and Rustad, S. A. (2008), A Sub-National Study of Child Soldier Recruitment in African Regions, Centre for the Study of Civil war (CSCW), PRIO, Norway.

Achvarina, V. and Reich, S. F. (2010), No Place to Hide: Refugees, Displaced Persons and Child Soldier Recruits, In: Gates, S. and Reich, S. (2010), Child Soldiers in the Age of the Fractured State, 1st Edition, University of Pittsburgh Press, USA.

Achvarina, V. and Reich, S. F. (2004), Why do children ‘fight'? Explaining child- soldiers ratios in intra-state conflicts, Ford Institute Working Papers, Ridgway Center for International Security Studies, University of Pittsburgh.

African Development Bank Group (2008), Democratic Republic of Congo, Result- Based Country Strategy Paper 2008-2012, January 2008.

African Development Fund (2004), Democratic Republic of Congo, Agricultural and Rural Sector Rehabilitation Support Project in Bas-Congo and Bandundu Provinces (Parsar), Agricultural and Rural Development Department Central and West Regions, Appraisal Report, March 2004.

Agborsangaya, O. (2000), Policy Dialogue to Advance the Process of Child Soldier Demobilization, in: Educational Assessment: Demobilization of Child Soldiers in the Democratic Republic of Congo, The Global Bureau Human Capacity Development Center, US Agency for International Development, August 2000, Washington DC.

Allard, P., Baxter, P., Halbwachs, M. and Komorowski J. C. (2002), Nyiragongo Volcano (Dem. Repub. Congo) and Related Hazards: Observations and Recommendations, Final Report of the French-British Scientific Team, March 8, 2002.

Ames, B. (2010), Methodological Problems In the Study of Child Soldiers, In: Gates, S. and Reich, S. (2010), Child Soldiers in the Age of the Fractured State, 1st Edition, University of Pittsburgh Press, USA.

104

Martin Sjøen Master Thesis University of Agder

Amnesty International (2003), Democratic Republic of Congo: Children at War, Al Index: Vol. 62, No. 34, 2003.

Andvig, J. C. (2006), Child Soldiers: Reasons for variation in their rate of recruitment and standards of welfare, Norwegian Institute of International Affairs, Norway.

Andvig, J. C. and Gates, S. (2010), Recruiting children for armed conflict, Centre for the Study of Civil War, Peace Research Institute Oslo, Norway.

Annan, J., Blattman, C., Dyan, M. and Kristopher, C. (2010), Civil War, Reintegration and Gender in northern Uganda, Journal of Conflict Resolution, Vol. 55, No. 6, pp. 877-908, December 2011.

Apio, E. (2008), Bearing the burden of blame – the children born of the Lord’s resistance army, northern Uganda, Coalition to Stop the Use of Child Soldiers. Available at: http://www.child-soldiers.org/psycho-social/english

Autesserre, S. (2007), D. R. Congo: Explaining Peace Building Failures, 2003-2006, Review of African Political Economy, Vol. 34, No. 113, pp. 423-441, 2007.

Azar, E. E. (1990), The Management of Protracted Social Conflicts, Dartmouth Publishing Co. Aldershot.

Baaz, M. A. and Stern, M. (2008), Making sense of violence: voices of soldiers in the Congo (DRC), Modern African Studies, No. 46, 2008, pp. 57-86, Cambridge University Press, United Kingdom.

Ballentine, K. and Nitzschke, H. (2005), The Political Economy of Civil War and Conflict Transformation, Berghof Centre for Constructive Conflict Management, 2005.

Barash, D. P. and Webel, C. P. (2009), Peace and Conflict Studies, 2nd Edition, SAGA Publications, USA.

Bayer, C. P., Klasen, F. and Adam, H. (2007), Association of Trauma and PTSD

105

Martin Sjøen Master Thesis University of Agder

Symptoms With Openness to Reconciliation and Feelings of Revenge Among Former Ugandan and Congolese Child Soldiers, American Medical Association, Vol. 298, No. 5, August 2007, USA.

Beber, B. and Blattman, C. (2011), The Logic of Child Soldiering and Coercion, Source: New York University/Yale University. Available at: http://chrisblattman.com/documents /research/2011.LogicOfChildSoldiering.pdf

Bell, M. (2006), Democratic Republic of Congo, Children Caught in War, UNICEF. Available at: http://www.unicef.org/childalert/drc/content/Child_Alert_DRC_en.pdf

Bell, R. H. (2007), Rethinking Justice, Restoring Our Humanity, Lexington Books, United Kingdom.

Blattman, C. (2009), From violence to voting: war and political participation in Uganda, American Political Science Review, Vol. 103, No. 2, pp. 231-247, USA.

Blattman, C. and Annan, J. (2008a), Child combatants in northern Uganda: Reintegration myths and realities, Security and Post-Conflict Reconstruction, Routledge, USA.

Blattman, C. and Annan, J. (2008b), On the nature and causes of LRA abduction: What the abductees say. In: Allen, T. and Vlassenroot, K. (Eds.). The Lord’s Resistance Army: War, Peace and Reconciliation in Northern Uganda, pp. 132-155, Zed Books, London.

Blattman, C. and Annan, J. (2007), The Consequences of Child Soldiering, Households in Conflict Network, The Institute of Development Studies, University of Sussex, Brighton.

Blattman, C. and Annan, J. (2010), The Consequences of Child Soldiering, The Review of Economics and Statistics, November 2010, Vol. 92, pp. 882-898, USA.

Blattman, C. and Miguel, E. (2010), Civil War, Journal of Economic Literature, Vol. 48, No. 1, pp. 3-57, USA.

106

Martin Sjøen Master Thesis University of Agder

Boothby, N. (2006), What happens when child soldiers grow up? The Mozambique case study, Intervention 2006, Vol. 4, pp. 244-259, No. 3, USA.

Borchgrevink, A. (2003), Silencing Language: Of Anthropologists and Interpreters, Sage Publications, Vol. 4, No. 1, 2003. Available at: http://eth.sagepub.com/content/4/1/95

Bouta, T. (2005), Assessment of the Ituri Disarmament and Community Reinsertion Program (DCR), Netherlands Institute of International Relations, Conflict Research Unit, May 2005, Netherlands.

Boya, L. K. (2010), D.R. Congo: The Darkness of the Heart, How the Congolese Have Survived Five Hundred Years of History, Xlibris Corporation, USA.

Boyden, J. (2010), Preface, In: Hart, J. (2010), Years of Conflict, Adolescence, Political Violence and Displacement, 2nd Edition, Studies in Forced Migration, Vol. 25, Berghahn Books, USA.

Boyden, J. and Berry, D. J. (2004), Children and the Youth on the Front Line, Ethnography, Armed Conflict and Displacement, Studies in Forced Migration, Vol. 14, Berghahn Books.

Brett, R. and Specht, I. (2004), Young Soldiers, Why They Choose to Fight, 1st Edition, Lynne Rienner Publishers Inc, USA.

Britannica (2011), Democratic Republic of the Congo, Britannica Encyclopedia. Available at: http://www.britannica.com/EBchecked/topic/132363/Democratic-Republic-of-the-Congo

Britannica (2012), Post-traumatic stress disorder (PTSD). Primary Contributor: Emery, Available at: http://www.britannica.com/EBchecked/topic/472284/post-traumatic-stress-disorder- PTSD/

Bryman, A. (2008), Social Research Methods, 3rd Edition, Oxford University Press, USA.

107

Martin Sjøen Master Thesis University of Agder

Cape Town Principles (1997), Cape Town Principles and Best Practices, UNICEF, 27- 30 April, Cape Town, South Africa.

Chikuhwa, T. W. (2010), The Evolution of the United Nations' Protection Agenda for Children: Applying International Standards, In: Gates, S. and Reich, S. (2010), Child Soldiers in the Age of the Fractured State, 1st Edition, University of Pittsburgh Press, USA.

CIA (2011), The Democratic Republic of Congo, CIA Fact-book. Available at: https://www.cia.gov/library/publications/the-world-factbook/geos/cg.html

Clark, J. F. (2002), The African Stakes of the Congo War, Palgrave Macmillan Publishing, USA.

Clément, C. (2009), Security Sector Reform in DRC: Forward to the Past, Security Sector Reform in Challenging Environments, Geneva Centre for the Democratic Control of Armed Forces (DCAF), Chapter 4, pp. 89-117, Geneva, Switzerland.

Coalition to Stop the Use of Child Soldiers (2008), Congo, Democratic Republic of the, Child Soldiers Global Report 2008, United Kingdom.

Coalition to Stop the Use of Child Soldiers (2011), Democratic Republic of Congo, Report to the Committee on the Rights of the Child in advance of the DRC initial report on the Optional Protocol to the Convention on the Rights of the Child on the involvement of children in armed conflict, April 2011, London.

Coalition to Stop the Use of Child Soldiers (2010), Mai Mai Child Soldiers Recruitment and Use: Entrenched and Unending, Briefing Paper: Democratic Republic of the Congo, February 2010, United Kingdom.

Coghlan, B., Brennan, R. J., Ngoy, P., Dofara, D., Otto, B., Clements, M. and Stewart, T. (2006), Mortality in the Democratic Republic of Congo: a nationwide survey, The Lancet, Vol. 367, pp. 44-51, January 7, 2006.

108

Martin Sjøen Master Thesis University of Agder

Collier, P. (2008) The Bottom Billion, Why the Poorest Countries Are Failing and What Can Be Done About It, Oxford University Press, New York.

Corbin, J. N. (2008), Returning home: resettlement of formerly abducted children in Northern Uganda, Disaster, Vol. 32, No. 2, pp. 316-336, USA.

Dagne, T (2011), The Democratic Republic of Congo, Background and Current Developments, CRS Report for Congress, Congressional Research Service, April 29, 2011, USA.

DAI (2007), Independent Evaluation of Special Projects for Child Soldiers in the Democratic Republic of Congo, Development Alternatives Incorporated, Presented to The World Bank’s Multi-country Demobilization and Reintegration Program (MDRP), February 5, 2007.

Dallaire, R. (2011), They Fight Like Soldiers, They Die Like Children, Arrow Books, Great Britain.

Deutsch, M. (1973), Conflicts: Productive and Destructive, In: Conflict Resolution Through Communication, Edited by Jandt, J. E. Harper and Row Publishing, New York.

Deutsch, M., Coleman, P. T. and Marcus, E. C. (2006), The Handbook Of Conflict Resolution, Theory and Practice, 2th Edition, Jossey-Bass Publishing, USA.

Dewalt, K. M. and Dewalt, B. R. (2002), Participant Observation, A Guide for Fieldworkers, Altamira Press, USA.

Druba, V. (2002), The Problem of Child Soldiers, International Review of Education, Vol. 48, No. 3-4, pp. 271-277, Netherlands.

Drønen, T. S. (2003), Afrikanske religioner, I Vi tror 2 – En innføringer i ulike religioner, naturreligioner, livssyn og trossamfunn, Tyri norsk forlag, Oslo.

Dunn, K. C. (2002), A Survival Guide to Kinshasa, in: Clark, J. F. (2002), The African

109

Martin Sjøen Master Thesis University of Agder

Stakes of the Congo War, Palgrave Macmillan Publishing, USA.

Eastern Congo Initiative (2011), Landscape Analysis of Community-Based Organizations: Maniema, North Kivu, Orientale and South Kivu Provinces of Democratic Republic of Congo, Full Report, Seattle.

Galtung, J. (1996), Peace by Peaceful Means, Sage Publications, London.

Gates, S. and Reich, S. (2010), Child Soldiers in the Age of the Fractured State, 1st Edition, University of Pittsburgh Press, USA.

Gates, S. and Reich, S. (2009), Think Again: Child Soldiers, What human rights activists never tell you about young killers, Foreign Policy, May 2009, USA. Available at: http://www.foreignpolicy.com/articles/2009/05/21/think_again_child_soldiers

Grossman, N. (2007), Rehabilitation or Revenge: Prosecuting Child Soldiers for Human Rights Violations, Georgetown Journal of International Law, No. 38, Vol. 323, Winter, 2007.

The Guardian (2012), Congo warlord Thomas Lubanga convicted of using child soldiers. Available at: http://www.guardian.co.uk/world/2012/mar/14/congo-thomas-lubanga-child-soldiers

Halvorsen, K. (2009), Å forske på samfunnet – en innføring I samfunnsvitenskapelig metode, 2nd Edition, Cappelens forlag, Norway.

Hart, J. (2008), Displaced children’s participation in political violence: Towards greater understanding of mobilization, in: Conflict, Security & Development, Vol. 8:3, pp. 277-293, Routlegde, USA.

Hart, J. and Tyrer, B. (2006), Research with children living in situations of armed conflict: concepts, ethics and methods, RSC Working Paper No. 30, Refugee Studies Centre, Oxford.

Hart, J. (2010), Years of Conflict, Adolescence, Political Violence and Displacement,

110

Martin Sjøen Master Thesis University of Agder

2nd Edition, Studies in Forced Migration, Vol. 25, Berghahn Books, USA.

Heckel, H. (2005), Transnational Advocacy against the Use of Child Soldiers and Sexual Exploitation of Children, In: Brem, S. and Stiles K. (2009), Cooperating without America: Theories and Case Studies of Non-Hegemonic Regimes, Routledge Publishing, USA.

Honwana, A. (2006), Child Soldiers in Africa, University of Pennsylvania Press, USA.

Humphreys, M. and Weinstein, J. M. (2004), What the Fighters Say, A Survey of Ex- Combatants in Sierra Leone June-August 2003, Interim Report: July 2004, USA.

IASC (2007), Guidelines on Mental Health and Psychosocial Support in Emergency Settings, Inter-Agency Standing Committee. Available at: http://www.humanitarianinfo.org/iasc/ content/products

ILO (2003), Wounded Childhood, The Use of Children in Armed Conflict in Central Africa, International Labour Office, April 2003, Washington DC.

International Peace Academy (2002), A Framework for Lasting Disarmament, Demobilization, and Reintegration of Former Combatants in Crisis Situations, IPA Workshop Report, German House, 12- 13 December 2002, New York.

IRC (2007), Mortality in the Democratic Republic of Congo: An Ongoing Crisis, The International Rescue Committee. Available at: http://www.peacebuildingdata.org/drc

Jackson, S. (2003), Fortunes of war, the coltan trade in the Kivus, HPG Background Paper Report 13, Overseas Development Paper, New York.

Johnson, K., Scott, J., Rughita, B., Kisielewski, M. Ashar, J., Ong, R. and Lawry, L. (2011), Association of Sexual Violence and Human Rights Violations With Physical and Mental Health in Territories of the Eastern Democratic Republic of Congo, The Journal of American Medical Association, Vol. 304, No. 5, pp. 553-562, August 2010.

111

Martin Sjøen Master Thesis University of Agder

Jolly, R. and Ray, D. B. (2006), The Human Security Framework and National Human Development Reports, United Nations Development Programme, Human Development Report Office, NDHR Occasional Paper 5.

Kanagaratnam, P., Raundalen, M. and Asbjørnsen, A. E. (2005), Ideological commitment and posttraumatic stress in former Tamil child soldiers. Scandinavian Journal of Psychology, Vol. 46, Issue 6, pp. 511-520.

Khan, R. E. (1999), Developing the Theoretical and Conceptual Framework, Mini- Lecture. Available at: http://journclasses.pbworks.com/f/theoretical+framework.ppt

Kimmerle, H. (2011), Ubuntu and Communalism in African Philosophy and Art, Published in: Heuvel, H., Mangaliso, M. and Bunt, L. (2006), Prophecies and Protest – Ubuntu in Glocal Management, Rozenberg Quarterly, The Magazine, 4. July, 2011.

King, G., Keohane, R. O. and Verba, S. (1994), Designing Social Enquiry, Scientific Interference in Qualitative Research, Princeton University Press, New Jersey.

King, G. and Murray, C. J. L. (2002), Rethinking Human Security, Political Science Quarterly, Vol. 116, No. 4, pp. 585-610, 2001-2002.

Kostelny, K. (2006), A culture-based integrative approach, Helping war-affected children, In: Boothby, N., Strang, A. and Wessells, M. (2006), A World Turned Upside Down, Social Ecological Approaches to Children in War Zones, Kumarian Press Inc, USA.

Koyame, M. and Clark, J. F. (2002), The Economic Impact of the Congo War, In: Clark, J. F. (2002), The African Stakes of the Congo War, Palgrave Macmillan Publishing, USA.

Lambert, A. and Lohlé-Tart, L. (2008), La surmortalité au Congo (RDC) durant les troubles de 1998-204: une estimation des décès en surnombre, scientifiquement fondée à partir des méthodes de la démographie, Unpublished report.

Lederach, J. P. (1997), Building Peace: Sustainable Reconciliation in Divided

112

Martin Sjøen Master Thesis University of Agder

Societies, United States Institute of Peace Press, Washington D. C.

Lee, A. J. (2009), Understanding and Addressing the Phenomenon of 'Child Soldiers', Working Paper Series No. 52, Refugee Studies Centre, Oxford.

Machel, G. (1996), Impact of armed conflict on children, UNICEF, New York.

McCalpin, J. O. (2002), Historicity of a Crisis, The Origins of the Congo War; In: Clark, J. F. (2002), The African Stakes of the Congo War, Palgrave Macmillan Publishing, USA.

McIntyre, A. and Weiss, T. (2003), Exploring Small Arms Demand, A Youth Perspective, Institute for Security Studies, Paper 67, March 2003, South Africa.

McMahan, J. (2010), An Ethical Perspective on Child Soldiers, In: Gates, S. and Reich, S. (2010), Child Soldiers in the Age of the Fractured State, 1st Edition, University of Pittsburgh Press, USA.

Mels, C., Derluyn, I., Broekaert, E. and Rosseel, Y. (2009), Screening for Traumatic Exposure and Posttraumatic Stress Symptoms in Adolescents in the War-Affected Eastern Democratic Republic of Congo, Archives of Pediatrics & Adolescent Medicine, Vol. 163, No. 6, pp. 525-530, June 2009.

Mels, C., Derluyn, I., Broekaert, E. and Rosseel, Y. (2010), The psychological impact of forced displacement and related risk factors on Eastern Congolese adolescents affected by war, Journal of Child Psychology and Psychiatry, Vol. 51, No. 10, pp. 1096-1104, October 2010.

Merriam-Webster (2012a), Hostility. An Encyclopedia Britannica Company. Available at: http://www.merriam-webster.com/dictionary/hostility

Merriam-Webster (2012b), Trauma. An Encyclopedia Britannica Company. Available at: http://www.merriamwebster.com/dictionary/trauma

113

Martin Sjøen Master Thesis University of Agder

Miall, H. (2004), Conflict Transformation: A Multi-Dimensional Task, 2nd Edition, Berghof Research Center for Constructive Conflict Management, Berghof.

Mitchell, C. (2002), Beyond Resolution, What Does Conflict Transformation Actually Transform? Peace and Conflict Studies, George Mason University, USA.

Morgan, D. L. (1997), Focus Groups as Qualitative Research, 2nd Edition, Sage Publications, USA.

Muggah, R., Maughan, P. and Bugnion, C. (2003), The Long Shadow of War: Prospects for Disarmament and Reintegration in the Republic of Congo, A Joint Independent Evaluation for the European Commission, UNDP and the MDRP Secretariat. Available at: www.oecd.org/dataoecd /49/32/35113279.pdf

Murison, J. (2002), The Politics of Refugees and Internally Displaced Persons in the Congo War, in: Clark, J. F. (2002), The African Stakes of the Congo War, Palgrave Macmillan Publishing, USA.

Orogun, P. S. (2002), Crisis of Government, Ethnic Schisms, Civil War, Regional Destabilization of the Democratic Republic of Congo, World Affairs, Vol. 165, No. 1, pp. 24- 41, 2002.

Paris Principles (2007), The Principles and Guidelines on Children Associated With Armed Forces and Armed Groups, February 2007.

Pauletto, E. and Patel, P. (2010), Challenging Child Soldier DDR Processes and Policies in the Eastern Democratic Republic of Congo, Journal of Peace, Conflict and Development, Issue 19, pp. 35.57, November 2010, London.

Peters, K. (2006), Footpaths to reintegration: Armed conflict, youth and the rural crisis in Sierra Leone, PhD Thesis, Wageningen University. Available at: http://library.wur.nl/WebQuery/ wda/lang/1804961

Peters, K. (2004), Re-examining voluntarism: youth combatants in Sierra Leone,

114

Martin Sjøen Master Thesis University of Agder

Institute for Security Studies, Pretoria.

Psychology Beyond Borders (2008), Psychosocial Adjustment and Social Reintegration of Children Associated with Armed Forces and Armed Groups: The State of the Field and Future Directions, Report prepared for Psychology Beyond Borders, USA.

Ramsbotham, O., Woodhouse, T. and Miall, H. (2008), Contemporary Conflict Resolution, 2nd Edition, Polity Press, Cambridge.

Renner, M. (1999), Arms control orphans, Bulletin of the Atomic Scientists, Sage Publications, No. 55, Vol. 1, pp. 22-26.

Reyntjens, F. (1999), The Second Congo War, More than a Remake, The Royal African Society, Oxford Journals, Vol. 98, No. 391, pp. 241-250, April 1999.

Rupesinghe, K. (Ed.) (1995), Conflict Transformation, Macmillan, London.

Rupesinghe, K. and Anderlini, S. N. (1998), Civil Wars, Civil Peace, An Introduction to conflict resolution, Pluto Press, London.

Rye, S.A. (2011), Research proposal and project planning, Mini-Lecture ME-413, Centre for Development Studies, University of Agder, Norway.

Schmidt, A. (2007), Volunteer Child Soldiers As Reality: A Development Issue For Africa, New School Economic Review, Vol. 2, pp. 49-76, 2007.

Schroeder, E. (2004), A Window of Opportunity in the Democratic Republic of Congo: Incorporating a Gender Perspective in the Disarmament, Demobilization and Reintegration Process. Available at: http://www.peacestudiesjournal.org.uk/dl/AWindowofOpportunity.PDF

Security Council (2010), Report of the Security Council mission to the Democratic Republic of the Congo (13 to 16 May 2010), United Nations, 30 June 2010.

Sendabo, T. (2004), Child soldiers: Rehabilitation and social reintegration in Liberia,

115

Martin Sjøen Master Thesis University of Agder

Life & Peace Institute Uppsala, Sweden.

SIK (2012), Fred og forsoning, Centre for Intercultural Communication. Available at: http://www.sik.no/?88

Singer, P. W. (2006), Children at War, 1st Edition, University of California Press, USA.

Singer, P. W. (2004), Talk is Cheap: Getting Serious About Preventing Child Soldiers, Cornell International Law Journal Vol. 37, pp. 561-586, 2004.

Singer, P. W. (2010), The Enablers of War: Causal Factors behind the Child Soldier Phenomenon, In: Gates, S. and Reich, S. (2010), Child Soldiers in the Age of the Fractured State, 1st Edition, University of Pittsburgh Press, USA.

Shepler, S. (2004), The Social and Cultural Context of Child Soldiering In Sierra Leone, Paper for PRIP Workshop Held in Oslo, August 20-21, 2004.

Somasundaram, D. (2002), Child soldiers: understanding the context, British Medical Journal, Vol. 324, No. 7348, May 2002, Sri Lanka.

Tefferi, H. (2010), Reconstructing Adolescence after Displacement: Experience from Eastern Africa, In: Hart, J. (2010), Years of Conflict, Adolescence, Political Violence and Displacement, 2nd Edition, Studies in Forced Migration, Vol. 25, Berghahn Books, USA.

Thakur, M. (2007), Demilitarising militias in the Kivus (eastern Democratic Republic of Congo), African Security Review 17.1, Institute for Security Studies.

Tonheim, M. (2010), Where are the research gaps? Reviewing literature on former girl soldiers, reintegration in the African context, in Mæland B. (ed.s) Culture, Religion, and the Reintegration of Former Child Soldiers in Northern Uganda, Geneva; New York: Peter Lang.

Tonheim, M. (2011), Who will comfort me? Stigmatization of girls formerly associated with armed forces and groups in eastern Congo, Centre for Intercultural Communication, Norway, Published by: The International Journal of Human Rights, Vol. 1-20.

116

Martin Sjøen Master Thesis University of Agder

Trefon, T. (2011), Congo Masquerade, The Political Culture of Aid Inefficiency and Reform Failure, Zed Books in association with the Royal African Society and Social Science Research Council, Great Britain.

UNDP (2011), Conge, Democratic Republic of the, International Human Development Indicators. Available at: http://hdrstats.undp.org/en/countries/profiles/COD.html

UNDP (2009), The Economic Crisis, Violent Conflict, and Human Development, A UNDP/ODS Working Paper, Office of Development Studies United Nations Development Programme, New York.

UNDP – DRC (2008), Exercise participatif d'analyse des conflits et capacités de paix pour la planification du dèveloppement dans la province du Nord-Kivu, Ville De Goma: Rapprt De Consultation, Republique Democratique Du Congo Ministere D'etat Charge De L'interieur Decentralisation Et Securite Province Du Nord-Kivu en collaboration avec Programme des Nations Unies our le Dévelopment Bureau de Goma, Nord-Kivu et Sud-Kivu, 4-5 Juin 2008.

UNESCO (2011), Ethical Guidelines for International Comparative Social Science Research in the framework of MOST, Social and Human Sciences. Available at: http://www.unesco.org/most/ethical.htm

UNICEF (2003), Guide to the Optional Protocol on the Involvement of Children in Armed Conflict, The United Nations Children’s Fund, December 2003, New York.

University of Minnesota (2003), A Guide To Research Ethics, Center for Bioethics. Available at: http://www.ahc.umn.edu/img/assets/26104/Research_Ethics.pdf

Utas, M. (2007), Building a future? The reintegration and remarginalisation of youth in Liberia, No Peace No War, An Anthropology of Contemporary Armed Conflicts, ed. Paul Richards, James Curry Ltd, Oxford.

Verhey, B. (2001), Child Soldiers: Preventing, Demobilizing and Reintegrating, Africa Region Working Paper Series, World Bank, No. 23.

117

Martin Sjøen Master Thesis University of Agder

Verhey, B. (2003), Going Home, Demobilising and reintegrating child soldiers in the Democratic Republic of Congo, Save the Children UK, February 2003, United Kingdom.

Vinck, P., Pham, P., Baldo, S. and Shigekane, R. (2008), Living With Fear, A Population-Based Survey On Attitudes About Peace, Justice And Social Reconstruction In Eastern Democratic Republic of Congo. Available at: http://www.peacebuildingdata.org/drc/about- drc-project/authors

Wessells, M. (2006), Child Soldiers: From Violence to Protection, Harvard University Press, Cambridge.

Wessells, M. (2010), Girls in Armed Forces and Groups in Angola: Implications for Ethical Research and Reintegration, In: Gates, S. and Reich, S. (2010), Child Soldiers in the Age of the Fractured State, 1st Edition, University of Pittsburgh Press, USA.

Wessells, M. and Monteiro, C. (2007), Psychosocial Intervention and Post-War Reconstruction in Angola: Interweaving Western and Traditional Approaches, Chapter 22, In: Christie, D. J., Wagner, R. V., and Winter, D. A. (Eds.). (2001), Peace, Conflict, and Violence: Peace Psychology for the 21st Century, Englewood Cliffs, New Jersey: Prentice-Hall.

Witzel, A. (2000), The Problem-Centered Interview, Forum: Qualitative Social Research, Sozialforschung, Vol. 1, No. 1, Art. 22, January 2000.

World-Gazetteer (2012), Congo, The Democratic Republic: largest cities and towns and statistics of their population. Congo (Dem. Rep.) cities. Available at: http://world- gazetteer.com/wg.php?x=&men=gcis&lng=en&des=wg&srt=npan&col= abcdefghinoq&msz=1500&geo=-46

Young, C. (2002), Contextualizing Congo Conflicts, Order and Disorder in Postcolonial Africa, In: Clark, J. F. (2002), The African Stakes of the Congo War, Palgrave Macmillan Publishing, USA.

Zimmermann, P. (2010), Interview: Congolese sanctuary offers child soldiers hope of readjustment, Deutsche Welle: Development Aid, June. 6, 2010. Available at: http://www.dw.de/dw/article/0,,5650813,00.html

118

Martin Sjøen Master Thesis University of Agder

Appendices

Appendix 1: Interview guide: In-depth interviews 120 Appendix 2: Overview of respondents: Background information 124 Appendix 3: Overview of collaborating organizations and institutes 125 Appendix 4: Ethical guidelines 126

119

Martin Sjøen Master Thesis University of Agder

Appendix: 1 Questionnaire for former child soldiers

Background information:

Name: ______Age: ______

Community: ______Date: ______

Pre-recruitment phase:

1. Where did you live before joining an armed group?

2. Who did you live with at that time?

3. What did you do (school/employment) before joining an armed group?

4. Did you live in poverty before you joined an armed group?

Recruitment phase:

1. How old were you when you joined an armed group?

2. How were you recruited?

- Were you voluntary of forcefully recruited?

- Why did you join an armed group?

3. Which armed group did you join?

4. How long were you a member of the armed group?

5. What role did you serve within the armed group?

- Have you experienced combat?

6. Why did you leave the armed group?

- Did you escape?

- Did someone negotiate for your release?

7. Have you been re-recruited since leaving the armed group?

120

Martin Sjøen Master Thesis University of Agder

- Were you voluntary or forcefully re-recruited?

- Which armed group re-recruited you?

8. What do you think is the reason for the armed conflict in DRC?

Reintegration:

1. What is your present living situation?

- Were you able to reunite with your family or relatives?

- How are you treated by your family and relatives now?

- Do you feel accepted by your family?

- Have you experienced any specific challenges when reuniting with your family and relatives?

- How do you think that these challenges should be dealt with?

- Has your living situation altered since you became reunited with your family?

- Has there been other positive or negative change?

- Are you satisfied living with your family or relatives?

- Do you wish you could live somewhere else?

2. Did you reintegrate into the same community as you lived in earlier?

- How are you treated by the local community?

- Do you feel accepted by your family?

- Have you experienced any challenges when reintegrating into the community?

- How do you think that these challenges should be dealt with?

- Has your living situation altered since for better or worse since you became reintegrated?

- Are you satisfied living in your local community?

- Do you wish you could live in another community?

121

Martin Sjøen Master Thesis University of Agder

3. What is your perception of hostility?

- Do you feel resentment towards any group or person?

- Have you used violence after leaving the armed groups?

- Have you experienced hostility from other people?

4. What is your current occupation?

- Are you skilled for a specific profession?

- Are you able to obtain a sufficient income from your work?

5. Have you partaken in purification rites or any other traditional ceremonies?

6. Have you received any offer to participate in intervention programmes?

- If not, would you have liked to receive such support?

Intervention program:

1. What made you join a formalized intervention program?

- Were you approached by someone?

2. What organization offered you intervention support?

- Have you received support from more than one organization?

3. Did you receive education, vocational training or employment assistance?

- What kind of livelihoods assistance did you receive?

- What is your impression of this support?

4. Did you receive any financial benefits or reintegration packages?

- What kind of financial benefits did you receive?

- Has it helped you in any way?

5. Have you received any follow-up support after being reintegrated?

- What kind of follow-up support have you received?

- Has it helped you in any way?

6. Were you able to make suggestions on what kind of assistance you would like?

122

Martin Sjøen Master Thesis University of Agder

- Was the assistance you received adapted to your individual needs and requests?

7. Do you believe that intervention support is beneficial for child soldiers?

Other issues:

1. What do you think about your future and what you want to do?

2. Is there anything else you would like to explain to me about your experience as a child soldier or your reintegration process?

123

Martin Sjøen Master Thesis University of Agder

Appendix: 2 Overview of respondents: Background information Participants Age Rec. Age Time served Group Recruitment Roles Exit mode Re-recruited Intervention Respondent 1 25 9 7 RCD Coercive Soldier Army Reunif. No Yes Respondent 2 24 17 4 n/a Coercive Auxiliary Escaped No Yes Respondent 3 32 16 6 months Interahamwe Coercive Soldier Escaped No No Respondent 4 19 17 3 Mayi-Mayi Coercive Soldier Army Reunif. Yes Yes Respondent 5 22 16 1 Interahamwe Coercive Auxiliary Escaped No Yes Respondent 6 27 12 4 RCD Coercive Spy Escaped No Yes Respondent 7 31 17 2 months Mayi-Mayi Coercive Auxiliary Escaped Yes No Respondent 8 28 15 3 Mayi-Mayi Voluntary Soldier Escaped No Yes Respondent 9 21 16 1 FARDC Coercive Auxiliary Army Reunif. No Yes Respondent 10 30 17 3 Interahamwe Coercive Auxiliary Escaped No Yes Respondent 11 22 n/a n/a Mayi-Mayi Coercive Soldier Escaped No No Respondent 12 18 12 2 CNDP Coercive Soldier Set free No Yes Respondent 13 20 n/a n/a Mayi-Mayi Voluntary Auxiliary Army Reunif. Yes Yes Respondent 14 27 17 2 FDLR Coercive Soldier Escaped Yes No Respondent 15 15 10 1 month Interahamwe Coercive Spy Escaped No Yes Respondent 16 17 11 4 CNDP Coercive Soldier Set free No Yes Respondent 17 21 16 3 RCD-ML Coercive Auxiliary Escaped No Yes Respondent 18 26 12 8 Interahamwe Voluntary Auxiliary Army Reunif. Yes Yes Respondent 19 22 17 1 FARDC Coercive Auxiliary Escaped No Yes Respondent 20 23 15 6 months Interahamwe Coercive Soldier Escaped Yes No Respondent 21 18 12 1 CNDP Coercive Auxiliary Escaped No No Respondent 22 23 11 n/a FDLR Coercive Spy Set free No Yes Respondent 23 19 15 3 Interahamwe Voluntary Auxiliary Escaped Yes Yes Respondent 24 25 14 1 Mayi-Mayi Coercive Soldier Escaped No Yes

124

Martin Sjøen Master Thesis University of Agder

Appendix: 3

Overview of collaborating organizations and institutes

Organization name Abbreviation Brief description Homepage Voluntary Force at the Service of Congolese non-profit NGO formed in Bukavu in 1992, http://www.bves- Childhood and Health BVES working with health, education and protection of children. rdc.org/DefaultEn.htm Centre for medical and psychosocial Congolese NGO providing psychosocial and medical assistance CAMPS treatment in South Kivu. n/a Norwegian humanitarian NGO working predominantly in Christian Relief Network CRN the Great Lakes Region. Active in DRC since 1994. http://crn.no/ German Agency for International Cooperation GIZ (GTZ) German development agency working in DRC since 1978. http://www.giz.de/en/ Prevention and reintegration of child Congolese NGO providing psychosocial and medical soldiers PREV treatment in South Kivu. n/a Norwegian Christian missionary organization working in Pentecostal Foreign Mission and Aid PYM DRC since 1922. http://pym.no/ Norwegian non-profit NGO established in 1986, RAFTO Foundation RAFTO dedicated to the promotion of human-rights. http://rafto.com/ Norwegian independent centre for research, teaching and Centre for Intercultural Communication SIK commissioned work, active in DRC since 2002. http://www.sik.no/

125

Martin Sjøen Master Thesis University of Agder

Appendix: 4 Ethical guidelines 1. Research participants must voluntarily consent to research participation and the degree of risk taken with research participants cannot exceed anticipated benefits of results. Human subjects must be allowed to discontinue their participation at any time.

2. Research must avoid unnecessary physical and mental suffering and no research pro- jects can go forward where serious injury and/or death are potential outcomes. Scien- tists must be prepared to terminate the experiment if there is cause to believe that con- tinuation will be harmful or result in injury or death.

3. Research should be conducted in such a way that the integrity of the research enter- prise is maintained, and negative after-effects which might diminish the potential for future research should be avoided.

4. The researcher should consider the effects of his/her work, including the consequences or misuse, both for the individuals and groups among whom they do their fieldwork, and for their colleagues and for the wider society.

5. The researcher should be aware of any potential harmful effects; in such circumstanc- es, the chosen method should be used only if no alternative methods can be found after consultation with colleagues and other experts. Full justification for the method cho- sen must be given.

6. The research should be carried out in full compliance with, and awareness of, local customs, standards, laws and regulations.

7. The principal investigators' own ethical principles should be made clear to all those involved in the research to allow informed collaboration with other researchers. Poten- tial conflicts should be resolved before the research begins.

8. Researchers are responsible for properly acknowledging the unpublished as well as published work of other scholars, as well as anyone providing assistance during this work.

9. Keep all names and information confidential. Never use any identifying names or pic- tures in written and recorded material.

10. The researcher must not raise expectations or make promises that cannot be fulfilled during the entire length of the fieldwork.

126

Martin Sjøen Master Thesis University of Agder

127